A Bipartisan Approach to Political Prosecutions (2024)

Abstract

1. INTRODUCTION

Politicians are seldom mistaken for saints. We suppose that they commit crimes no less than ordinary citizens. Boss Tweed, Spiro Agnew, Richard Nixon, Edwin Edwards, Kwame Kilpatrick—this list could go on and on. All agree that politicians should be held to account for their offences. Pols are not above the law. Like the rest of us, they are under it.

The difficulty is that prosecutors are often politicians themselves, eager for renown, and greater power. Prosecutors routinely run for attorney general, governor, Senator, or even President. Hence, prosecutors sometimes have motivated, personal reasons to hound the opposition and thereby play to portions of their partisan base. If a prosecutor sends a prominent opposition pol to jail, their co-partisans will fete them and back their bids for higher office. If the prosecution fails, the prosecutor may nonetheless be given partial credit for attempting to hold their opponents to account. Many of their co-partisans may imagine that the opposition figure was guilty, and that legal technicalities or irresponsible jurors thwarted justice.

The tension between accountability and politicization has never been more acute. Three prosecutors have brought four prosecutions against former President Donald Trump across four jurisdictions. Manhattan District Attorney Alvin Bragg brought state law charges against Mr. Trump and secured a conviction on all 34-counts (District Attorney Bragg Announces 34-Count Felony Indictment of Former President Donald J. Trump 2023). In Florida and Washington, D.C., Special Counsel Jack Smith pursues federal charges, one related to government records and one related to Trump’s attempts to delay the Congress’s certification of the 2020 election results (Tracking the Trump Criminal Cases 2023). Finally, state District Attorney Fani Willis is prosecuting the former President, along with a host of others, in Fulton County, Georgia (McWhirter et al. 2023).

Elite reactions varied across these indictments. With respect to Jack Smith’s prosecution in Florida, Democrats have been mostly supportive and some Republicans have been dismissive. But there are exceptions on the Right. William Barr, the former Attorney General, said that the Florida obstruction indictment is “very strong, because a lot of the evidence comes from his own lawyers” (Transcript: Former Attorney General William Barr on “Face the Nation” 2023). Further the “legal [defense] by which he [Trump] gets to take battle plans and sensitive national security information as his personal papers is absurd” (Transcript: Former Attorney General William Barr on “Face the Nation” 2023). Such comments perhaps reflect the relative strength of this indictment.

The responses to the Washington D.C. indictments have been mixed. Some believe that the D.C. charges are the most significant because they relate to Trump’s bid to retain office. Many Democrats feel that Trump attempted to undermine democracy and feel vindicated by these charges. And yet some Democrats have admitted that this case “is more challenging” because it turns on Donald Trump’s intent and because some of these charges rest on novel legal theories (Mariotti 2023). The Republican reaction has been more critical. Some Republicans believe that the charges reflect a desire to hound a declared candidate for President. But even the most thoughtful Republican critics of Trump, such as Jack Goldsmith, worry that the charges were brought too late, rest on speculative legal theories, and raise tricky issues of intent (Goldsmith 2023).

In contrast, the initial reactions to the New York state prosecution were disquieting. From the left, there was a sense of overreach and dismay. The Washington Post Editorial Board—hardly a pro-Trump outfit—published an opinion piece called the New York charges “shaky” and the legal strategy “novel” (Editorial Board 2023a).1 Similarly, a left-leaning commentator said that the indictments rested on a “dubious legal theory” (Millhiser 2023).

Public perceptions are united in some respects and polarized in others. The unity is paradoxically troubling, for many agree that the sky is falling. Seventy-one percent of Republicans believe that the “politicization of law enforcement” is a very important or somewhat important issue in determining how they will vote (Reuters/Ipsos Poll 2023). Seventy-three percent of Democrats agree, perhaps because they worry that their leaders will be prosecuted next (Reuters/Ipsos Poll 2023). The disunity is more frightening still. Among Democrats, ninety-one percent believe that the Department of Justice is holding Trump accountable (Weiner et al. 2024). But most Republicans and many Independents see it differently. Seventy-seven percent of Republicans and forty-three percent of Independents believe that prosecutors have “unfairly targeted Trump for political reasons” (Weiner et al. 2024). These are astoundingly high numbers.

To some, the guilty verdicts from the New York trial may suggest that initial concerns about politicization were overblown. Yet some insist that appellate courts may overturn these convictions on one or more grounds (Honig 2024). And others say the prosecution should never have been brought, whatever its legal merits. For instance, after the guilty verdicts, Former New York Governor Andrew Cuomo asserted that the prosecution should never have been brought and would not have been had the target not been Donald Trump (Nazarro 2024). The implication is that the New York prosecution was manifestly political. One former federal prosecutor said the following: the “charges against Trump aren’t just unusual. They’re bespoke, seemingly crafted individually for the former president and nobody else” (Honig 2024).

We take no position on whether Bragg should have prosecuted Trump, much less the merits. Furthermore, we seek neither to attack nor defend Trump. Instead, we are focused on bigger questions. These prosecutions and the varied reactions to them underscore the difficulties of prosecuting prominent politicians. Some see overdue justice in the making and are gratified, bordering on exuberant. Others see a form of corruption—the misuse of the machinery of the state to harass the opposition. Some such critics will be disillusioned and others will be incensed.

The latter reaction is particularly corrosive, for it spurs cries for retribution. Mike Davis, of the Article III Project has argued that “Republicans need to learn to take off the gloves” and open up investigations into the Bidens (Nelson 2023a). If Alvin Bragg “can turn a routine settlement of a business dispute seven years ago into a felony, I think our Republican AGs and DAs should get creative,” and go after Biden (id.). “Two wrongs don’t make it right, but it makes it even” (id.). Judicial Watch’s President Tom Fitton has stated that he believes there are prosecutors in Florida and Texas who are “looking for state law hooks into the Biden family,” and if they are not, “they’re not doing their jobs” (Nelson 2023b). When one of us was quoted as saying that any politicized prosecutions brought by a future Trump administration would usher in a “banana republic,” (Arnsdorf et al. 2023) emails poured in. One citizen said the nation “already DOES resemble [a banana republic.] Democrats targeting their top opponent with 91 spurious indictments is not a good look” (Email from Anonymous Sender 2023). He is not alone, as the polling signals.

Republicans critical of Trump have made alarming predictions of revenge. Vin Weber predicted that “[w]e’re going to see political prosecutions brought, some of them for meritorious reasons, some of them to advance the careers of the prosecutors. But all of this is harmful to America and our political process” (Bolton 2023). Barr, someone who turned on Trump, believes that if the latter returns to the White House, he will politicize prosecutions to “go after his enemies” (Wolf 2023). Goldsmith predicts that we might witness “ever more aggressive tit-for-tat investigations of presidential actions” and an “exacerbate[ion of] the criminalization of politics” (id.).

Progressives can see the storm on the horizon, the payback in the offing. Some are saying a second Trump term would be a “revenge presidency” (Ignatius 2023). Last year, the former President publicly said that he may prosecute Joseph Biden and his family via a special counsel (Samuels 2023), a remarkable comment. This year, he said “Joe would be ripe for Indictment. By weaponizing the DOJ against his Political Opponent, ME, Joe has opened a giant Pandora’s Box” (Concepcion 2024).

The former President is not off the mark in observing that the Pandora’s Box is now open. If you disagree with Trump, as many might, you may well believe that he will open the Box, should he become President. Given the high percentage of Republicans who believe that Trump is being victimized, it makes it easy for a future Trump Justice Department to prosecute prominent Democrats. Republicans are primed to conclude that any prosecution of Joseph Biden, or even Jack Smith, is entirely justifiable under the law and others will believe that it is a fitting payback. In an almost perfect mirror image, the Republican base will see those future prosecutions precisely the way that many Democrats today regard every Trump prosecution. Even if Trump never returns to the Oval Office, at some point there will be revenge prosecutions. Aggrieved Republicans are not likely to hold their fire and wrath and will, sooner or later, respond in kind.

In short, we are on the verge of a new reality where future federal prosecutors and some of the 2330 state prosecutor offices (Perry & Banks 2011, p. 1) will feel empowered, even obliged, to prosecute federal officials with whom they disagree politically. We can even imagine a world in which politicians avoid traveling to jurisdictions controlled by political rivals for fear of state prosecution. Partisan righteousness and grievances plus the personal opportunism of prosecutors make it open season on the political opposition.

The Supreme Court’s recent decision endorsing a measure of criminal immunity for Presidents does little to address well-founded fears about politicization. In fact, Trump v. United States reflects an underlying concern about the politicization of prosecutions. The Court concluded that Presidents were absolutely immune with respect to the exercise of “core”—exclusive and preclusive—constitutional powers (Trump v. United States, No. 23-939, slip op. at 6 (US July 1, 2024). It further concluded that other presidential actions—constitutional and statutory—are at least presumptively immune and some of them might be absolutely immune ( Id., p. 6, 14). While this framework will provide a good deal of protection, it is hardly an impenetrable shield. First, prosecutors may bring politicized charges against a President grounded on his non-official acts (including acts undertaken when the President was not in office) (Id., p. 15). Second, prosecutors may prosecute based on the president’s official acts that are outside of the “core” (Id., p. 9); whether they will succeed is another question. Third, and relatedly, prosecutors continue to have electoral incentives to prosecute even when there is a strong chance that a court will dismiss the prosecution before trial. Prosecuting a former President will endear them to a portion of their partisan base even if a court ultimately says that the underlying acts are immune from prosecution. And, of course, the decision supplies no protection to Senators, judges, presidential candidates, etc. The opinion is narrowly focused on presidents and former presidents. In contrast our proposal covers many more prominent federal officials and office seekers.

Further, the Court’s opinion reflects a certain unease with the many prosecutions brought against Trump, or so we suppose. The Court says that without immunity there is a serious risk that a new administration will prosecute a former president (Id.,p. 40–41). Perhaps the Court says this because it imagines that this is what happened to Trump, at least in part. Here is the prediction of a vicious cycle: “[T]he more likely prospect [is] an Executive Branch that cannibalizes itself, with each successive President free to prosecute his predecessors” (Id., p. 40). The Court then cited a vague statute, section 371, that makes it a crime to engage in a conspiracy to impair, obstruct, or defeat “the lawful function[s]” of the federal government (Id.). “An enterprising prosecutor... may assert that a previous President violated that broad statute” by failing to enforce certain laws related to immigration, etc. (Id.) “Without immunity, such types of prosecutions of ex-Presidents could quickly become routine”.(Id.) In forecasting retribution, the Court perhaps imagined that the many investigations and prosecutions of Trump reflected a modicum of partisanship.

Returning to the particulars of the Trump prosecutions, the actions that have brought us to the precipice, our sense is that while some of them seem strong, others seem weak. Bringing weak prosecutions against politicians does little to bolster confidence in the American government. Rather, they sap it at every turn. The targeted politician draws a terrible lesson—politics is a no-holds barred slugfest, where one can, and should, weaponize the state against opponents. Their supporters draw the same lesson—we cannot allow the other side to target ours without retaliation in kind. In contrast, bringing strong prosecutions—where there is something of a bipartisan consensus that the case was warranted—can perhaps yield greater confidence in the system. As the Wall Street Journal has said, “an indictment of a former President should be based on serious charges with enough evidence to convince most Americans that it is justly brought” (Editorial Board 2023b). When that happens, people may have more faith in their government.

The Department of Justice is aware of the need to assuage the public. As one DOJ report puts it, the Department’s actions must have “the appearance and the reality of fairness and impartiality” (Annual Report 2019). When prosecuting politicians, the DOJ must steer clear of the Scylla of an acrimonious prosecution and the Charybdis of a timid choice to look the other way and do nothing. In a hyper-partisan era, the Department has an unenviable task.

To mitigate concerns, any federal prosecution of a federal politician must be widely perceived as grounded in expertise and judgment, and not in partisanship or prosecutorial self-interest. Given pervasive and legitimate concerns about the weaponization of prosecutions, there is a need for a process to allay public fears. We propose a mechanism meant to better insulate federal prosecutions from politics.

The solution lies in imposing an innovative check on decisions to prosecute pols. Ordinarily, the criminal process consists of a grand jury indictment followed by a trial before a petit jury. We propose an additional step between indictment and prosecution: the federal prosecution of a federal politician should be brought only if the DOJ first secures the consent of a Prosecutor Jury—a jury composed of former US Attorneys. The Department would randomly pull jurors from a pool of former US Attorneys, until 10 nominated by Republican Presidents and 10 nominated by Democrats agreed to serve. Once the Prosecutor Jury is seated, DOJ prosecutors would present evidence and seek the panel’s consent to prosecute. If more than two-thirds (at least 14) of the 20 prosecutors endorsed a prosecution, a federal prosecutor could prosecute.

Our two-thirds requirement deliberately replicates key supermajoritarian features of the Constitution. Two-thirds of the Senate must vote to convict before someone is removed from office.2 Likewise, two-thirds of a chamber must vote to expel a member.3 As we have seen with the impeachments of Bill Clinton and Donald Trump, this super-majority requirement means that removal will occur only when House prosecutors can convince a substantial minority of Senators from the defendant’s own party. The same was true of the recent expulsion of Republican George Santos, where a significant number of his co-partisans had to vote to expel Santos (Floran & Talbot 2023). When it comes to impeachment or an expulsion from a chamber, a partisan process will invariably fail.4

Our proposal extends beyond former Presidents and encompasses former Cabinet members, current members of Congress, and federal judges, each of whom who are vulnerable to the same abuses. Hence, our proposal shields the likes of Hakeem Jeffries, Chuck Schumer, Sonia Sotomayor as well as Mitch McConnell, Mike Johnson, and John Roberts. Additionally, it would also cover major-party nominees for President, the House, and the Senate. Because our proposal extends to incumbent legislators, it should garner support from both sides of the congressional aisle. Seasoned politicians understand that what goes around can come around, for today’s wrongful tormentors may become tomorrow’s woeful targets. Relatedly, some federal legislators may especially welcome the extra protection from federal prosecutions.

The DOJ might voluntarily impose this check upon its prosecutors. Alternatively, Congress could mandate the use of Prosecutor Juries in federal prosecutions. Furthermore, states could adopt comparable procedures to address similar concerns with potential partisanship in state prosecutions. We are confident that state Prosecutor Juries would have blocked the lawless 2014 indictment of Texas Governor Rick Perry (Malewitz & Ramsey 2016) but would have sanctioned the indictment of Illinois Governor Rod Blagojevich (Jeong 2021).

Our framework helps ensure that politicians would not be above the law because they remain subject to prosecution. But politicians and, more importantly, the nation, would be better protected from partisan prosecutions because Prosecutor Juries would have members attuned to any potential weaponization of the DOJ. Should a Prosecutor Jury recommend a prosecution, Americans would know that a supermajority of professionals endorsed the prosecution, and that half of the jurors were appointed by Presidents of the opposite party.

Some may protest that our reform gives politicians extra protections not enjoyed by ordinary Americans. But our framework is not in the service of the political class. Rather the procedures ensure that one set of politicians does not abuse another cohort. The Blackstonian ratio, that “[i]t is better that ten guilty persons escape than that one innocent suffer” (Blackstone 1765a–1769a, 352)—applies with equal force when elected officials are being prosecuted. Indeed, it might have greater force. In the context of prosecuting high officials, when a good portion of the country perceive that a politician is facing a political vendetta, democracy suffers. Without the support of at least a substantial minority of the defendant’s party there is a severe risk that prosecution will be perceived as partisan. In these times of tribalism, the American Republic must not devolve into a banana republic where criminal prosecutions are used as a cudgel against political opponents.

Part 2 describes the risk of partisan prosecutions and why the current Justice Department procedures concerning politically motivated prosecutions are insufficient. Part 3 describes the Prosecutor Jury and explains why it provides a useful prophylactic. Part 4 reports the results of two surveys of former US attorneys, where, as a proof of concept, we try to produce simulacrums of a Prosecutor Jury, albeit one focused on indictments.5 While there are several important caveats, the results are nonetheless illuminating.

We believe our proposal is worthy of consideration however the Trump prosecutions ultimately fare. In any case, a focus on Trump deranges the discussion, preventing some from recognizing the problems with the current regime and how our framework mitigates them. For a moment, imagine an alternative history, one where in 2017 the Trump Department of Justice had prosecuted Hillary Clinton for her allegedly illegal handling of classified materials. Or envision that Trump wins in 2024 and his DOJ prosecutes Joe Biden and his brother James Biden. Obviously, millions would regard such prosecutions as grounded in rank partisanship. A Prosecutor Jury could blunt such concerns by preventing a prosecution. Alternatively, a Prosecutor Jury might lend credence to any prosecution. In any event, our principal point is that both parties can weaponize prosecutions. And, importantly, both parties can be perceived as doing so. We seek to combat the reality and perception that prosecutors might be biased against prominent opposition figures. Hence, we believe it vital to devise a mechanism to head off (or defuse) claims that future prosecutions are witch hunts of political opponents.

2. THE PROBLEM OF PARTISAN PROSECUTIONS

In deciding whether to prosecute politicians, the Justice Department must navigate two vital and seemingly irreconcilable imperatives: first, politicians should not be immune from criminal prosecution, and second, the decision to prosecute should not be influenced by partisan politics or be seen to be so motivated. A prosecution should be a vehicle to punish lawbreakers, but it should not be driven by an impetus to punish political adversaries. The Department of Justice has a host of rules designed to prevent partisan considerations from playing a role in investigations and prosecutions. Unfortunately, as discussed below, the existing rules do not adequately preclude inferences that prosecutorial decisions are motivated by partisanship.

2.1 The Twin Requirements of Politically Sensitive Prosecutions

It is a bedrock principle of the rule of law that even the most powerful political actors are bound by legal constraints and are subject to legal consequences when they fail to honor these constraints. In Britain, the King could do no wrong (Blackstone 1765b–1769b, p. 237) and hence was not subject to any legal process, much less criminal trial and punishment. In America, Lex is to be Rex and not even the President is above the law. As future Supreme Court Justice James Iredell would say in 1788, “No man is better than his fellow-citizens, nor can pretend to any superiority over the meanest man in the country. If the President does a single act by which the people are prejudiced, he is punishable himself” (Elliott 1836, p. 106, 109).6 If that is true of the President, it is likewise true for other federal officials and mere candidates. Similarly, in 1883, the Supreme Court opined:

No man in this country is so high that he is above the law. No officer of the law may set that law at defiance with impunity. All the officers of the government, from the highest to the lowest, are creatures of the law and are bound to obey it (United States v. Lee, 106 U.S. 196, 220 (1882)).7

Under this republican principle, all officials, former and current, are under the law and may be criminally prosecuted for their wrongdoings.8

Yet subjecting high officers, current and former, to criminal prosecution opens the possibility that prosecutors will wield their considerable powers to harass and harm their political adversaries. Further, members of the public might infer, rightly or wrongly, that prosecutors are persecuting opponents.

Why would prosecutors misuse their prosecutorial authority? All officials face the temptation to exploit their authority for petty reasons, to feel mighty, or to lord it over others. Even the lowest government functionary may wish to make citizens grovel before them or make them squirm. But there are other reasons why prosecutors might abuse their authority. In America, prosecutors are political creatures, at least in the sense that they are appointed or elected via a political process. At the federal level, US Attorneys are appointed by the president, typically with the Senate’s advice and consent (28 U.S.C. § 541).9 This means that they are often quite loyal to the incumbent, and some may feel pressure to protect the President and impede or harass his opponents.

Further, prosecutors are often ambitious. Everyone knows that many prosecutors have designs on elected office, federal or state, with their tenure in the prosecutor’s chair serving as a lofty platform from which to seek a still higher perch. Thomas Dewey ran for governor of New York, and then for President, with his exploits as a special prosecutor serving as a feather in his cap (The Dewey Story 1948). Vice President Kamala Harris cited her extensive prosecutorial experience in her many campaigns (My Record as a Prosecutor: Kamala Harris—NAACP 2019). Successfully prosecuting a presidential candidate, a former President, or a Senator will add luster to a resume, making it easier to run for Governor, Senator, or President.

In 1940, the Attorney General, Robert Jackson, candidly discussed the potential for prosecutorial abuse:

The prosecutor has more control over life, liberty, and reputation than any other person in America. His discretion is tremendous. He can have citizens investigated and, if he is that kind of person, he can have this done to the tune of public statements and veiled or unveiled intimations. Or the prosecutor may choose a more subtle course and simply have a citizen’s friends interviewed. The prosecutor can order arrests, present cases to the grand jury in secret session, and on the basis of his one-sided presentation of the facts, can cause the citizen to be indicted and held for trial (Jackson 1940, p. 3).

Needless to say, these troubling tactics can be inflicted upon opposition politicians.

Given legitimate concerns about the potential wrongful conduct of prosecutors, judicial doctrine has erected a series of safeguards. The Supreme Court has held that vindictive prosecutions violate the requirement of due process (Bordenkircher v. Hayes, 434 U.S. 357 (1978); see alsoNewland & Parker 2020). Furthermore the Court has said that a federal prosecutor’s overarching duty is not to “win a case, but [insure] that justice shall be done” (Berger v. United States, 295 U.S. 78, 88 (1935)). Prosecuting someone because of their political views or actions flouts these principles.10 Politically motivated prosecutions not only subject the defendant to harassment and unwarranted intrusions they also injure the electorate and undermine confidence in our system of government.

2.2 The Department’s Rules

The Justice Department has additional, internal procedures that seek to guard against politically -motivated investigations and prosecutions. There are a series of rules that forbid certain improper considerations in investigations and prosecutions and, in particularly sensitive cases, insist upon consultation with the upper echelon of the DOJ, including the Attorney General.

The Department’s Justice Manual expressly bars consideration of a person’s “political association, activities, or beliefs” in “determining whether to commence or recommend prosecution or take other action against a person” (U.S. Dep’t of Just., Just. Manual § 9-27.230 (2018)). The Department reinforces these substantive limitations with some procedural safeguards,11 ones that require the participation of the Department’s Public Integrity (PIN) Section of the Criminal Division in some politically-sensitive investigations. The PIN Section “has exclusive jurisdiction over allegations of criminal misconduct on the part of federal judges and also supervises the nationwide investigation and prosecution of election crimes” (About the Public Integrity Section 2023).

According to a 2019 annual report, the PIN Section’s work principally “focuses on public corruption, that is, crimes involving abuses of the public trust by government officials” (Annual Report 2019, p. 1)12 These corruption cases “tend to raise unique problems of public perception” and are “often politically sensitive because their ultimate targets tend to be politicians or government officials appointed by politicians” (Annual Report 2019, p. 1). The Section has emphasized that:

A successful public corruption prosecution requires both the appearance and the reality of fairness and impartiality. This means that a successful corruption case involves not just a conviction but public perception that the conviction was warranted, not the result of improper motivation by the prosecutor, and is free of conflicts of interest (Annual Report 2019, p. 1–2).

While this admonition on its face only involves corruption charges, the points are generalizable. Any prosecution of a high federal official, including office seekers, should have the “appearance and the reality of fairness and impartiality” (Annual Report 2019, p. 1–2).

Indeed, Section 9-85.500 of Department’s Justice Manual reflects a keen and sound awareness that the investigation or prosecution of a politician raises extraordinary concerns. Adopted in 2022, this provision requires consultation with the PIN Section before taking many politically sensitive actions that might influence an election:

Federal prosecutors and agents may never select the timing of any action, including investigative steps, criminal charge, or statements, for the purpose of affecting any election, or for the purpose of giving an advantage or disadvantage to any candidate or political party.... Any action likely to raise an issue or the perception of an issue under this provision requires consultation with the Public Integrity Section, and such action shall not be taken if the Public Integrity Section advises that further consultation is required with the Deputy Attorney General or Attorney General (U.S. Dep’t of Just., Just. Manual § 9-85.500 (2022)).

Further this provision not only requires that the PIN Section be heard on these issues, but also gives the PIN Section the power to force review by superiors (Shortell 2020).13 In a sense, the Section has a say in any prosecution of any politician, federal or state, even if the potential charges do not relate to public integrity. For instance, if a politician is accused of drug use and there may be a “perception” that any investigation or prosecution will be seen as political or partisan, the Section should be involved, with the possibility of participation by the highest echelon of the Department of Justice.

The DOJ Justice Manual provides additional protections to members of Congress, creating restrictions that apply without regard to election season. Section 9-85.110 provides:

Consultation with the Public Integrity Section is required in all investigations involving a Member of Congress or congressional staff member. In particular, the Public Integrity Section must be consulted prior to taking any of the following steps: (1) interviewing a Member of Congress or congressional staff member; (2) subpoenaing a Member of Congress or congressional staff member; or (3) applying for a search warrant for a location or device in which legislative materials are likely to be found. In addition, consensual monitoring in an investigation involving allegations of misconduct by a Member of Congress requires approval by a Deputy Assistant Attorney General (U.S. Dep’t of Just., Just. Manual § 9-85.110 (2018)).

Requirements of consultation and high-level approval can help ensure that political or personal bias do not influence an investigator’s decisions to probe members of Congress and their staff.

In 2020, William Barr adopted additional election-related rules about candidates for the presidency and Congress, along with their campaigns and staff. In particular, “no investigation... may be opened or initiated by the Department” without the prior consultation of an Assistant Attorney General and the relevant US Attorneys (Barr 2020, p. 2). The hurdle is higher for presidential and vice presidential candidates, their campaigns, and their staff, because any such investigation also requires the “written approval” of the Attorney General (Barr 2020, p. 2). Like many of the rules discussed above, the Barr Memorandum focuses on opening an investigation and therefore might be seen as particularly focused on the Federal Bureau of Investigation (FBI) and not on prosecuting attorneys. One suspects that the rule reflects Barr’s disquiet with the FBI’s investigations of Hillary Clinton and Donald Trump in the lead up to the 2020 election. Barr has criticized the latter investigation (Zapotosky & Barrett 2019).

As the 2022 election season approached, Attorney General Merrick Garland endorsed the Barr Memorandum’s rules. But Mr. Garland went further because his memo also explicitly addressed prosecutions. He declared that “partisan politics must play no role in the decisions of federal investigators or prosecutors regarding any investigations or criminal charges” (Garland 2022, p. 1). Henceforth every DOJ officer and employee must consult with the PIN Section when they “face an issue, or the appearance of an issue regarding the timing of statements, investigative steps, [or] charges... near the time of a primary or general election” (Garland 2022, p. 1). Presumably, the Chief of the PIN Section would speak to the Assistant Attorney General in charge of the Criminal Division, who might rope in the Deputy Attorney General and the Attorney General.

While many of the rules discussed above are meant to ensure supervision by the highest echelons of the DOJ, another set of laws and rules are better seen as insulating decision-making from the influence wielded by high-level departmental officials, particularly the Attorney General. In the wake of Watergate, Congress enacted the Ethics in Government Act of 1978 (Pub. L. No. 95-521, 92 Stat. 1824 (1978)), one of whose provisions provided for the appointment of a tenure-protected Independent Counsel to investigate high executive branch officials (Pub. L. No. 95-521, § 601, 92 Stat. 1824, 1867 (1978)). Under that act, famous independent prosecutors pursued allegations of wrongdoing against Secretaries, Attorneys General, and Presidents. Lawrence Walsh and Kenneth Starr were the most prominent, but hardly the only such prosecutors.

With the expiry of those provisions in 1999, Attorney General Janet Reno created internal rules for the appointment of special counsels who would enjoy tenure protection via regulation (28 C.F.R. § 600 (2019)). These rules remain in force. They leave the Attorney General considerable discretion as to whether and when to appoint a special counsel (28 C.F.R. § 600.1 (2019)). The Attorney General may appoint a special counsel when there is a “conflict of interest” within the DOJ, where the “public interest” suggests a need, or where “extraordinary circ*mstances” exist (28 C.F.R. § 600.1 (2019)). In other words, the Attorney General has unreviewable discretion over whether to appoint a special counsel.

Once appointed, however, special counsels are meant to enjoy a measure of independence from the Department. To be sure, special counsels must follow extant departmental policies, procedures, and rules (28 C.F.R. § 600.7(a) (2019)). Nonetheless, they wield extraordinary power. Special counsels enjoy the “full power and independent authority to exercise all investigative and prosecutorial functions of any United States Attorney” (28 C.F.R. § 600.6 (2019)). Further, they exercise considerable latitude, because the Attorney General has explicitly barred their “day-to-day supervision” (28 C.F.R. id. § 600.7(b) (2019)). This implies that any Departmental review is, at best, intermittent or irregular (28 C.F.R. id. § 600.7(b) (2019)).

Crucially, the only person who may remove or discipline a special counsel is the Attorney General (28 C.F.R. id. § 600.7(d) (2019)). Removal may occur only for “misconduct, dereliction of duty, incapacity, conflict of interest, or for other good cause,” including violation of Departmental policies (28 C.F.R. id. § 600.7(d) (2019)). Although a seemingly long list, these sorts of constraints, when enacted into law, have long been thought to greatly constrain the President’s removal authority. Besides the substantive removal standard, there is an implicit requirement of a judicial or quasi-judicial proceeding prior to removal. As Professor Aditya Bamzai has noted, similar statutory standards have been understood to demand a judicial proceeding involving evidence, witnesses, and an opportunity for a hearing (Bamzai 2018). Needless to say, any such hearing poses a daunting task for the Attorney General, for a high threshold for removal is backed by a deliberate, quasi-judicial process.

In sum, once the Attorney General appoints a special counsel, the counsel’s relative independence from the DOJ is meant to assure the public that any investigation or prosecution is not the result of pressure from the Attorney General or the White House. Special counsels are a species of independent counsels, albeit with slightly less tenure protection.

Special counsels are feared creatures. When Rod Rosenstein appointed Robert Mueller as special counsel to investigate the Russian collusion theory, Donald Trump apparently said, “Oh my God. This is terrible. This is the end of my Presidency. I’m f*cked” (Key Excerpts from the Mueller Report 2019). Similar thoughts must have gone through Trump’s head with the appointment of Jack Smith. Relatedly, we imagine that Merrick Garland’s appointment of Robert Hur to investigate Joseph Biden’s handling of classified documents (Appointment of a Special Counsel 2023) was most unwelcome news in the White House and in the Oval Office in particular.

When one considers all these DOJ rules together, they seem to reflect the principle that elected officials (and those campaigning for high office) should neither be above nor below the law. Elected officials and candidates should be susceptible to indictment, trial, and punishment for crimes that they commit. Furthermore, no one, candidate, incumbent, or otherwise, should be targeted for their political beliefs or affiliation. There are also some internal mechanisms, in the form of consultation with, and approval by, higher ups that apply to any investigation of a member of Congress and any investigation of a politician that raises potential concerns about the perception of politicization. These rules are designed to ensure that lower-level officials are not conducting malicious investigations and to ensure that any investigations are conducted with sensitivity to public perceptions. Somewhat paradoxically, other rules seek to insulate certain investigations and prosecutions from such supervision, to ensure that the Attorney General and their minions will not tightly oversee and direct an investigation and prosecution.

2.3 The Current Rules are Inadequate

As needful as they are, these Departmental rules are insufficient, particularly at this moment of tribalism. There has always been a problem of the politicization of the criminal process, or at least the perception of politicization. But the tribalism of modern politics, coupled with certain politicians enjoying a cult-like following, has made the problem manifestly worse.

At the outset, it is useful to recount that Presidents have long been accused of persecuting opponents. In the wake of the Whiskey Rebellion, there were Americans who believed that the use of state militias to suppress the tax uprising was a plot to suppress liberties and silence the voices of dissenters (Chervinsky 2020, 262–263 (describing James Madison and Thomas Jefferson as critics)). This perception regarding the politicization of law enforcement, and the reality, went from bad to worse. President John Adams used to read newspapers, identify seditious writings, and then direct his district attorneys to prosecute the authors for violations of the Sedition Act (Prakash 2005, 1743–1745). Adams believed that he was protecting the government from malicious attacks. His successor, Thomas Jefferson, directed politically-sensitive prosecutions as well, including his supervision of the Aaron Burr treason trial (Prakash 2005, 1747–1748). The former Vice President was Jefferson’s rival, having refused to concede the election of 1800.

The politicization has not ended. Laurence Walsh was seen by Republicans as pursuing a partisan investigation of Reagan-era officials (Lamar Jr. 1988). Kenneth Starr was seen by Democrats as transforming an investigation into a land deal into an investigation of President Bill Clinton’s sexual peccadillos (Gugliotta 1998). James Comey was seen by Democrats, and then by Republicans, as engaged in a political witch hunt of each party’s standard bearer (Haberman et al. 2017; Hulse 2017(describing Democrats’ frustration towards James Comey during Hillary Clinton’s emails investigation); (describing Republicans’ frustration towards Trump during the Russia-Trump collusion investigation)). Whether the investigators, or the critics, were politicizing the investigations is irrelevant for our purposes. The fact is that all these investigations were perceived by a good number of Americans as political and partisan.

In a reversion to older practices, modern Presidents are again wading into investigations and prosecutions. In the 2016 campaign, Donald Trump initially reveled in and encouraged the chants of “Lock Her Up,” a reference to the jailing of Hillary Clinton for mishandling classified documents (Wright & Wolf 2016). Trump’s stance could be seen as a pledge that should he become president, he would lock her up. After he won the election, he said that the Hillary Clinton matter should be laid to rest (McCaskill & Nelson 2016). But he later toyed with the idea that she ought to be prosecuted, particularly when he was in legal peril. In 2018, President Trump shared a meme of his political opponents pictured in a jail cell, including Hillary Clinton.14 Further, there were reports that President Trump wanted his administration’s opponents prosecuted (Weiser 2022). Trump has time and again pronounced Biden guilty of “treason” and other unspecified crimes, calling for him to be sent to prison (Benen 2020; Mallin 2020; Squire & Griffiths 2024). He did this while President and he did so after he lost the election (Benen 2020; Mallin 2020; Squire & Griffiths 2024).

President Joseph Biden has been more reticent to comment, and his interventions have been less bold. Yet he too has said things that might influence investigators and investigations and hence has contributed to the perception that they might be politicized. Before the 2020 election, his campaign characterized the Hunter Biden Laptop story as Russian disinformation (Cheney & Bertrand 2020), a claim that was designed to influence the election but might have influenced investigators as well. With respect to his son, Hunter Biden, the President has repeatedly said that his son violated no law. For instance, Hunter did not violate the income tax laws (Mangan 2020), said Biden. He has also effusively praised his son and said he is proud of him (Liptak & Perez 2022), something one expects of a father. Yet when a President publicly declares that he is the proud father of a target of multiple criminal investigations and that his son is innocent, such pronouncements may curb the willingness of investigators and prosecutors to pursue the target, particularly where the investigators and prosecutors owe their jobs, in some measure, to the President.

With respect to former President Trump and his mishandling of classified documents, President Biden said in an interview that he was not monitoring the investigation’s progress because he did not want to influence it. Yet he also publicly said that he wondered “how that [the retention of classified documents] could possibly happen” and “how anyone could be that irresponsible” (Pelley 2022). He went on: “I thought, ‘What data was in there that may compromise sources and methods?’” Biden further said. “It’s just totally irresponsible” (Pelley 2022). Given that the crime involves the mishandling of classified documents, these comments create the distinct perception that Biden believes that Trump is guilty. In any event, one can create a perception of improper influence without any real intent to influence. Further, one can intentionally influence investigators without saying anything directly to them. While Trump’s numerous interventions were hardly subtle, some will regard Biden’s statements as improper, even if far less egregious.

When Biden found himself the subject of a special counsel investigation, one also related to classified documents, he had to field questions. In response to reporters, he said “I think you’re going to find there’s nothing there. I have no regrets” (Cannella 2023). He also said that “there’s no ‘there’ there” (Pelley 2022). On the one hand, targets of investigations typically say that they have done nothing wrong. On the other hand, targets of investigations typically are not the President of the United States, with authority and influence over the investigators and prosecutors. We will likely never know whether his pronouncements influenced the investigation.

In this cauldron of politics and tribal allegiances to one party or figure, how are we to evaluate the Department’s rules about prosecutions not being influenced by partisanship, bias, etc., and, importantly, not being perceived as being under such influences? Again, we believe the existing rules and structures, while laudable, do not meet the moment. They are deficient, relying too much on admonitions and too little on more meaningful structural reform.

To begin with, while many prosecutors will honor rules barring the use of partisan affiliation or political beliefs in prosecutorial decisions, others may not. In other realms, we have rules backed by mechanisms for enforcement and sanctions for violations. Officers take an oath to the Constitution.15 But they can be prosecuted for treason when they take certain actions that transgress that vow.16 Similarly, prosecutors take an oath to enforce the law.17 But there are other mechanisms and procedures in place to ensure their adherence to law. To punish someone, prosecutors must go to a grand jury, and then argue before a jury and judge (Steps in the Federal Criminal Process 2024). Moreover, they can be punished if they corruptly influence investigations and proceedings.18 We do not say that because they take an oath to follow the law, we are confident that they invariably will honor, and not pervert, the law.19

Furthermore, the use of grand and petit juries does not wholly protect against the wrongs of partisan prosecutions. It is widely understood that grand juries tend, in the words of Sol Wachtler, the former Chief Judge of New York, to “operate more often as the prosecutor’s pawn than the citizen’s shield” (Levin 2014 (“[D]istrict attorneys now have so much influence on grand juries that ‘by and large’ they could get them to ‘indict a ham sandwich.’”)). There is a stronger likelihood that a petit jury might prevent a wrongly-prosecuted politician from being convicted—especially because a single-juror can prevent the required unanimity for conviction. But forcing wrongly-accused politicians to figuratively stand in the dock and defend themselves is too great a cost to their time, resources, and reputation. More importantly, politically-motivated prosecutions weaken our democracy even if they end in hung juries, for a good portion of the public will be co*cksure that the prosecution was biased and the result of some partisan plot. The sooner a partisan prosecution ends, the better off society is for it.

The Special Counsel rules are another mechanism meant to instill public confidence. Those rules are, in part, designed to ward off charges that Departmental personnel are favoring certain targets and disfavoring others. By appointing a special counsel, the Attorney General seems to hand off control of a criminal inquiry to an official who is meant to be, in some measure, independent of the incumbent administration. Hence when a special counsel investigates the President’s political opponent, the hope is that the counsel’s measures will be seen as non-partisan.

But these rules do not work well. Many special counsels do not inspire public confidence. To the contrary, their actions sometimes suggest a crusade or vendetta against their targets.

In an ideal world, special counsels would investigate a targeted official and decide, based solely on the merits, whether prosecution and punishment are warranted. Special counsels would feel perfectly free to decline to bring charges, just as thousands of prosecutors do each day. All prosecutors must make judgments about where to allocate resources and which alleged offenders are worth pursuing and which are not. Most of these decisions merit little notice or comment because they are not politically charged.

But every special counsel faces a different set of incentives and expectations. To begin with, a special counsel’s huge resources and relatively narrow jurisdiction create conditions where counsels are apt to find wrongdoing as a means of justifying their extraordinary appointment. If you create a special counsel for a single, average citizen, the counsel would likely find some wrongdoing because, after all, there are many laws and many opportunities for wrongdoing, either purposeful or inadvertent, and because special counsels want to show something for all their exertions. Scholars face the same temptation when they expend a good deal of effort researching something—reading, thinking, surveying, experimenting. After exerting themselves, academics are prone to finding or discovering “remarkable” and “novel” results. There are few or no rewards for undertaking a prodigious investigation and then discovering nothing.

This tendency is perhaps exacerbated because of the spotlight on special counsel investigations. Special counsel investigations of cabinet officers, Presidents, and political opponents are the most sensitive and fraught investigations precisely because of their political ramifications. Investigate a sitting President and there are millions of people with huge stakes in the outcome. The stakes are no less important when you investigate the President’s potential opponent in the general election.

Occasionally, the targets of investigations, and their prosecutors, have exacerbated the political nature of these investigations. Bill Clinton (and his allies) sharply attacked Ken Starr (Baker & Schmidt 1998). Donald Trump (and his allies) sharply attacked Robert Mueller (Smith 2019). Starr’s and Mueller’s independence from the Department of Justice did little to shield them from such attacks. Moreover, in both cases, the prosecutors were accused of being partisan. In Starr’s case, he employed many attorneys associated with the Republican party (Fisher 1998). Mueller likewise employed many lawyers tied to the Democratic party (Gregorian 2019).

Hence, while special counsel investigations perhaps insulate the Department of Justice (and its leadership) from charges of partiality, they do little to counter accusations of partisanship and bias. Because special counsels often claim to have unearthed wrongdoing, they invariably inflame supporters of the accused. These supporters quickly surmise that a special counsel’s investigation and prosecution is a product of bias and partisanship. We see such assertions, once again, in the context of Donald Trump. Every special counsel that brings a prosecution is seen as pursuing a wrongful vendetta in furtherance of their own fame and career.

3. THE PROSECUTOR JURY

Given the inadequacy of existing Department of Justice procedures meant to avoid the appearance and reality of politicized prosecutions, an additional screen is necessary. We propose a novel check on DOJ decisions to prosecute politicians holding or pursuing federal office.

The Department should randomly select a politically-balanced group of 20 former US attorneys to form a Prosecutor Jury. The DOJ prosecutor should supply the evidence and the relevant charges that were presented to the grand jury or that were discovered via the grand jury proceedings. The Prosecutor Jury would evaluate the evidence and the DOJ’s legal claims. If two-thirds of the prosecutors (14) on the Prosecutor Jury approved of the charges, the Department could prosecute those charges approved by the Grand and Prosecutor Juries.

Adding a Prosecutor-Jury screen can help ensure that any prosecution does not merely reflect improper partisan motivations. Former US Attorneys who were nominated by Presidents of the same political party as the potential defendant are likely to be particularly attuned to the possibility of improper motivations. Because prosecutions can only proceed to the trial stage if at least a third of attorneys nominated by the same party concur, the process sends a clear message to the public that the Department was able to persuade a substantial minority of those who likely share a party affiliation with the accused. The process responds to the legitimate concern that a prosecutor may prosecute an opposition figure, or rival office seeker from the same party, even when the evidence is thin or the legal theory for some or all the counts is underwhelming.

Our proposal usefully harnesses or exploits partisanship in the service of bipartisanship and consensus on prosecutorial decisions. First, we recognize that prosecutors can be motivated by bad motives, personal or political. We counter that tendency by using other prosecutors with different political affiliations as a partial check on a prosecutor acting with excessive zeal or improper motives. The process generates prosecutorial decisions that are more rock-solid and more justifiable. Second, the kind of vituperative political divisions that makes partisan prosecutions a greater possibility in the modern era also make our proposal more politically feasible. Both Democrats and Republicans have reason to fear that they will become targets if a President of the opposition party comes to power. It is easier to sleep at night if you know that you can be indicted only when a substantial minority of co-partisans agree. Hence, we believe that there might be great support for our reform amongst members of Congress.

In this section, we discuss details of implementation and defend the Prosecutor Jury against several potential criticisms. But we pause at the outset to emphasize several problems that are proposal does not mitigate, much less address. First, though investigations can be a product of partisan motivations (or be seen as a product of them), our proposal does not impede the politicization of investigations. For instance, our proposal would not have prevented the FBI’s public intervention in the 2016 race. Under existing rules, DOJ investigators would still be required to consult with the Public Integrity Section with regard to politically-sensitive investigations. But to the extent that this process fails to preempt partisanship, officials might use investigations to harass and embarrass political opponents. These forms of overstepping are left unaddressed in part because there is an additional social value in the possibility that investigations will bring information to light that would be relevant to voters, and in part because it is infeasible to stop inappropriate investigations without also chilling the ability of appropriate investigations to proceed. Further, whether (and how) an investigation proceeds will turn on highly particularized judgments, ones not susceptible to useful second-guessing by outsiders. For all these reasons, we do not propose any new rules for commencing and continuing investigations.

Second, we do not propose any mechanism whereby a prosecutor’s decision to decline to prosecute would be second-guessed. A choice not to prosecute could be controversial. For instance, James Comey’s conclusion that Hillary Clinton committed no crime was divisive (Parks et al. 2017). But we think reform should be directed to what seems the most consequential decision, namely a choice to utilize the machinery of government to commence a prosecution against a politician. Besides, the executive has an implicit constitutional right to decline to bring charges,20 a claim backed by the awesome authority to pardon before trial.21 In other words, there is no lawful way for Congress to order the executive branch to prosecute, either using a Prosecutor Jury or otherwise.

Third, our proposal does not combat the many other ways by which prosecutors may politicize their decision-making. It seems to us infeasible to have something like a Prosecutor Jury intrude into the myriad pretrial and trial decisions of prosecutors. How to present the facts, which witnesses to call, which objections to raise, what to say in court—these should be left to a single person, advised by others. Prosecution by committee seems inadvisable.

Fourth, our mechanism provides less protections for potential defendants of fringe-parties or for defendants who have substantially different politics than the US Attorneys of the same political party. A Prosecutor Jury may provide little additional protection for fringe-party defendants, whose party is unappealing to former prosecutors of either party. Then again, one might argue that politically motivated prosecution of fringe-party members does less democratic damage because these individuals were, by definition, less likely to be viable contenders for office.

Fifth, our framework pays no heed to how parties evolve over time and how US Attorneys appointed in the more distant past might have views and outlooks that greatly diverge from more recent appointees. For instance, pre-Trump Republican nominees likely harbor less solicitude for President Trump. President Trump is quick to label any Republican who criticizes him as a RINO (Republican in name only).22 While we take no position on who deserves to call themselves a true Republican, it is possible that US Attorneys nominated by Ronald Reagan, George H.W. Bush, or George W. Bush may be biased against President Trump in ways that reduce their ability to serve as a prophylactic against a partisan prosecution. Indeed, one could imagine some of them believing that the Republican party, or the nation, would be better off if Trump were in jail.

An alternative proposal might require some minimum representation of recent US Attorneys and demand that a substantial minority of them support prosecution before the DOJ could proceed. We think either prerequisite is unwarranted. First, both create too great a possibility that a group of nominated loyalists would thwart an appropriate prosecution. Second, both rules are unwieldy and unworkable as applied to other politicians. For instance, should a Senator facing prosecution be able to insist that the Prosecutor Jury contain a minimum number of US Attorneys that she voted to confirm because she is part of movement to transform her party? We favor a Prosecutor Jury Pool that reflects a party’s historical successes and evolution over time and not one that gives greater weight to the party’s latest trends and fixations.23

Finally, while our proposal mitigates the possibility of politicized prosecutions, we admit that it cannot ensure that Americans will no longer regard certain prosecutions as political. Even after a Prosecutor Jury approves of some charges, the targeted politician will continue to claim that prosecutors are victimizing them. Relatedly, some Americans will take their cues from their favored pols rather than objectively assess the signal that emerges from a Jury’s decision to sanction the prosecution. Nonetheless, even if not every member of the public is swayed by the Prosecutor Jury’s decisions, the Jury supplies objective evidence of bipartisan support that gives the best chance of influencing members of the public that are persuadable.24 Over time, the public is likely to see that Prosecutor Juries make nuanced decisions. For example, bipartisan support for prosecuting George Santos and Robert Menendez is more likely than bipartisan support for prosecuting Hillary Clinton or Scooter Libby (Cabral 2024 (“[Menendez] has resisted bipartisan pressure to resign from the Senate”); Freking 2023). Persuading even a portion of the public is worth something, for it will decrease the tendency to see charges (or the absence of charges) as biased and wrongful. If fewer Americans see a politicized prosecutorial machine, that is a rather welcome outcome.

3.1 The Details of Implementation

The Department of Justice would compile a pool of all living former US Attorneys. Persons within the Prosecutor Jury Pool are those who have enjoyed the confidence of the President, the Attorney General, and, one might say, the Department of Justice. For each member of this Pool, there is a strong presumption of experience, integrity, and professionalism.

When a federal prosecutor wishes to prosecute a covered politician, she would first obtain a grand jury indictment. If the prosecutor secured an indictment, the DOJ would empanel a Prosecutor Jury. As noted, the Department would randomly select US Attorneys from the Pool until it had twenty former prosecutors willing to serve on the Prosecutor Jury. Randomly choosing from the hundreds of former US Attorneys25 would help ensure the appearance of representation from a broad swath of both political parties.26 Because we seek partisan balance, ten of the jurors would be US Attorneys appointed by Democratic Presidents and ten would be US Attorneys appointed by Republican Presidents. We suggest that existing recusal mechanisms, both voluntary and involuntary, ought to apply to all US Attorneys selected to serve on the Jury.27 Further, we suppose that some potential jurors who hold state or federal office would be precluded from serving on the jury pool. The remaining jurors would principally be retired US Attorneys or former US Attorneys who are in private practice, with a firm or otherwise.

The actual prosecutor would present evidence and lay out the charges to the empaneled Prosecutor Jury. The evidence would have come from the investigation or the grand jury proceedings. The charges would have come from the grand jury. At the end of that presentation, the Jurors would consider the evidence and the law and make a professional judgment about whether a prosecution would be appropriate. If two-thirds of the Jurors approved of one or more charges, the federal prosecutor could then prosecute. The rest of the criminal process would play out as it currently does.

A Prosecutor Jury might approve of all the charges in the proposed prosecution. This would, no doubt, gratify the prosecutor. Alternatively, the Jury might reject every one of the prosecutor’s recommendations. This would strongly signal that the prosecutor had overreached. Perhaps the most intriguing and useful possibility is that the Jury approves some, but not all, of the charges, an outcome that suggests the Prosecutor Jury has performed a rather vital function. Winnowing out the charges would underscore that only the most credible criminal counts were brought to trial. The American people should have great confidence in the charges that make it through this filtration process.

An additional benefit arises from the nature of the participants. Unlike the grand jury or the petite jury, the Prosecutor Jury is an expert panel. Former US Attorneys better appreciate what the alleged facts show and do not show. They understand federal law far better than most laypeople. They recognize which charges, if any, ought to be brought and which counts are stretches or legally inappropriate. They even have a sense of appropriate exercises of prosecutorial discretion and when charges might be legally permissible but nonetheless inadvisable. Finally, as compared to grand and petite juries, the Prosecutor Jury is unlikely to be led around by the nose or to be deferential to the prosecutor seeking to prosecute.

We propose that there be a separate Prosecutor Jury for each politician that a prosecutor wishes to indict. So, if prosecutors sought to prosecute five politicians in related cases, or otherwise, there would be five separate Prosecutor Jury panels. Of course, an alternative would be to create a handful of standing Prosecutor Jury panels, each of which would serve a year. This would mirror the term of a grand jury. It would mean that members of each Jury panel would be agreeing to serve for a year and hence would constitute a greater burden on their time.

Grand jury proceedings are sealed, meaning that only the people in the room know what was said before, and to, the grand jury. We would replicate some of these secrecy rules for the Prosecutor Jury. The discussions and proceedings should be kept confidential to protect the Prosecutor Jurors, some of whom may wish to deliberate in complete privacy. Yet the public ought to know which grand jury indictments the Prosecutor Jury rejected. Specifically, after a grand jury publicly indicts, it would be useful for the public to know which charges the prosecutor brought to the Prosecutor Jury and which of those the Prosecutor Jury rejected.28 This would make the Jury’s filtration function crystal clear.

Below, we discuss five additional implementation issues.

3.1.1 Who Should Impose Or Adopt Prosecutor Juries?

Reform can come from within or without. Sometimes reform arises internally and organically, reflecting a widespread, in-group sense that some change is necessary in order to improve a public institution’s standing with the people. Other times, outsiders—typically legislators—must impose reform because the institution is unable (or refuses) to see the need.

Both the 2020 Barr memorandum and the 2022 Garland memorandum are, in our view, useful and sincere attempts to remind Department personnel to eschew partisan bias and to avoid taking steps that influence elections. The Department of Justice must not become the Department of Election Influence. We obviously think more needs to be done to avoid that possibility.

Because Attorneys General possess legal authority to seek out advice and create additional rules guiding prosecutions,29 any Attorney General might voluntarily summon a Prosecutor Jury in any case where they believe it would be useful, whether a potential prosecution involved politicians or not. The option of using a Prosecutor Jury might serve as a substitute for the appointment of a special counsel. Further, the Department could voluntarily, by rule, impose a Prosecutor Jury requirement over a subset of important federal politicians or candidates for office.

Of course, lawmakers might enact a law to impose a Prosecutor Jury requirement. They might begin by protecting a small set of covered individuals and later decide to expand the class to cover themselves and others. Because Congress can regulate the means and mechanisms of federal law execution it can impose this procedural constraint on executive decisions to prosecute.

While we encourage the DOJ to adopt Prosecutor Juries as a best practice, we worry that any internal guidelines would be susceptible to executive override in times of pressure and crisis. For example, if the DOJ had voluntarily adopted Prosecutor Juries during the Trump presidency, the President might have nonetheless directed the Justice Department to indict Hillary Clinton without empaneling a Prosecutor Jury. Department officials could resign, but, as with the Saturday Night Massacre (Elving 2018), a President hellbent on upending an internal guideline is likely to find an official who, for institutional reasons or otherwise, is willing to do the executive’s bidding.

In contrast, if the requirement of empaneling a Prosecutor Jury for certain politically sensitive prosecutions was enshrined in a statute, then the President would be unable to sidestep the requirement—just as an executive cannot evade the requirement of a grand jury’s approval.30 If the Department of Justice, at the President’s behest or otherwise, tried to bypass the Prosecutor Jury mandated by law, a defendant could raise it as a defense at trial and seek to have the indictment quashed. Hence even as we favor the DOJ voluntarily adopting the Prosecutor Jury, we also favor Congress incorporating the framework into a law.

We also propose that states adopt analogous versions of our proposals to cover state prosecutions of both federal and state officials. The criminal prosecutions of former President Trump by state prosecutors in New York and Georgia make clear that the appearance of politically-motivated prosecutions does not merely arise with respect to federal prosecutions. An incumbent politician or a candidate should not be immune from liability for violating state criminal law; but neither should they be subject to overreaching by a plethora of state prosecutors that might have ulterior motives in prosecuting.31

3.1.2 How Binding Should The Decisions Of Prosecutor Juries Be?

The ultimate decisions of a Prosecutor Jury could be made more or less binding on prosecutors. For example, a decision that the Department should not prosecute could be made to be merely advisory—analogous to advisory jury judgements in certain criminal and civil contexts (Holtzman 2021; Ronner 1995). If the Prosecutor Jury’s recommendation against some charge was made public, the Department might feel sufficiently restrained and prosecute only in extraordinary circ*mstances. While reasonable people might disagree on this point, we prefer a more robust role for the Prosecutor Jury, one where its decisions are more than precatory. When the Prosecutor Jury fails to sanction a charge, the Department should have no power to prosecute the politician for the alleged violation.32

3.1.3 Which potential defendants should be covered?

Though the investigation of former Presidents and candidates for president is our central motivating example, the problem of partisan prosecutions extends beyond these targets. Political calculations, of various sorts, might cause federal prosecutors to inappropriately target prominent officials, such as members of Congress or federal judges. We have already cited the wrongful state prosecution of high-level officials, including Governor Rick Perry (Malewitz & Ramsey 2016). But the problem extends to federal prosecutors as well. The famed prosecution of Aaron Burr, who had become an adversary of President Thomas Jefferson, ended with the jury finding him not guilty (Bomboy 2023). The prosecution of Governor John Connolly in the “Milk Money” scandal fizzled out with a not guilty finding (Nation: The Milk Case Revisited 1979). More recently, the infamous prosecution of Senator Ted Stevens ended in a spectacular failure,33 perhaps suggesting that the prosecution ought not to have been brought in the first instance. Finally, whatever one thinks of the current prosecution of Senator Robert Menendez , a previous prosecution ended in a mistrial followed by a DOJ decision to drop all charges (Barrett 2018).

A vital question in considering the appropriate breadth of coverage is estimating how many Prosecutor Jury pools would be formed each year. The 2021 Report of the Public Integrity Unit reveals that every year between 600 and 1100 federal, state, and local officials were charged with public corruption offenses (Annual Report 2021, p. 25). In our opinion, this many uses of the Prosecutor Jury is unwarranted, suggesting that the list of covered officials ought to be smaller.

We recommend the use of a Prosecutor Jury whenever a federal prosecutor seeks to prosecute anyone who, within last five years, (i) has held elected federal or statewide office, (ii) has received more than 20% of the vote in a general election contest, federal or statewide, (iii) has held a Senate-confirmed office, or (iv) has held a federal judgeship.34 We have intentionally included those especially vulnerable to politically-motivated prosecutions. These coverage rules would encompass members of Congress, federal judges, and high-ranking executive officials, including cabinet-members, military officers, and other high executive officers. They also would cover some unsuccessful office seekers. Finally, they cover prominent statewide figures, including governors.

Beyond this list, we think that the Prosecutor Jury might usefully cover certain relatives of politicians. For instance, we might include the immediate family of some of the most prominent covered individuals, each of whom might be a plausible target of partisan prosecutions. The immediate relations—spouse, parents, siblings, children—of former Presidents and candidates for President perhaps ought to be covered. Neither Hunter Biden nor Donald Trump Jr. should be targeted because of their relationship to a prominent politician (Khardori 2022; Sheil 2022; Thomas 2022; Totenberg 2017). Nor should Melania Trump or Jill Biden. But, of course, the broader the protections afforded to relatives of politicians, the more costly our proposal becomes.

Even this wide list may leave out some significant people who are arguably targeted because of their political affiliations. Roger Stone, for example, falls outside the set of covered individuals.35 To deal with such situations, decisionmakers within the DOJ or the Congress might consider giving a leader of each party the power to confer this protection on ten people, with the safeguard extending for ten years. The sitting President would be the natural leader of one party, and the implementing rule or statute might give the designation power to the House Speaker if they are of a different party than the President, or, if the same party, to the House minority leader.

Admittedly, reasonable people will have different intuitions about coverage. Some will favor a narrower ambit, one that perhaps extends only to Presidents, presidential candidates, congressional leadership (e.g. House Speaker and Senate majority leader), and federal judges. They might suppose that the problem is not as widespread as we have suggested and that the pool of potential jurors is rather limited. These are valid points, ones that the ultimate decisionmakers, in the Department of Justice or Congress, would have to carefully weigh. In any event, we welcome those who see merit in our general proposal even as they differ as to the ideal coverage.

3.1.4 Why Former US Attorneys?

Which politically associated individuals are best suited to serve on a Prosecutor Jury? We have chosen former US Attorneys for a number of reinforcing reasons. First and foremost, they have relevant experience and expertise. As US Attorneys, they had to make difficult charging decisions. And they have long had to make such decisions while thinking about the public consequences of their prosecutorial decisions. As noted earlier, they have a comparative advantage compared to grand jurors and petite jurors because each Prosecutor Juror is a lawyer who can distinguish legal stretches from solid legal claims. Second, they tend to be politically astute. Most navigated an executive and legislative process in being nominated by the President and confirmed by the Senate for their positions. And their appointment as US Attorneys is often not the culmination of their public careers. The list of former US Attorneys includes luminaries who have gone on to hold other federal posts, e.g. Samuel Alito, Richard Blumenthal, Jeff Sessions, James Comey, and Sheldon Whitehouse.36

3.1.5 What Questions Should the Prosecutor Jury Answer?

We think that the Prosecutor Jury should decide whether a prosecution is warranted. But it should do more. To begin with, the Prosecutor Jurors should implement the Justice Department guidelines for “Commencing or Declining Prosecution.” These require the prosecutor to believe that “the person’s conduct constitutes a federal offense, and that the admissible evidence will probably be sufficient to obtain and sustain a conviction” (U.S. Dep’t of Just., Just. Manual § 9-27.220 (2023)). Prosecution, however, would be unwarranted where “(i) the prosecution would serve no substantial federal interest; (ii) the person is subject to effective prosecution in another jurisdiction; or (iii) there exists an adequate non-criminal alternative to prosecution” (U.S. Dep’t of Just., Just. Manual § 9-27.220 (2023)). Prosecutor Jurors should be able to second guess a charge on any of these grounds.

More generally, we think that Prosecutor Jurors should be free to second-guess the policy judgment reflected in a decision to prosecute. In particular, the Jurors ought to be able to consider whether the federal interest substantially advanced by a prosecution is outweighed by other national interests that counsel against going forward with a prosecution. Those interests might include whether confidence in our democracy or the nation’s international standing counsels against proceeding against a politician. Another way of putting the point is that the prosecution decision is inevitably an all-things considered decision and in second-guessing a prosecutor, the Prosecutor Jurors ought to be able to consider all the factors they believe are relevant, subject of course, to whatever restrictions that the DOJ normally imposes upon prosecutors.37

Finally, and relatedly, Prosecutor Jurors should be admonished to honor the Department guideline to act without any “purpose of giving an advantage or disadvantage to any candidate or political party” (U.S. Dep’t of Just., Just. Manual § 9-85.500 (2022)). While we recognize that such admonitions will not invariably preclude improper motives, they are useful on the margins.

3.2 Defending the Prosecutor Jury

Critics of our reform might argue it is unnecessary, unconstitutional, and violates the rule of law. We disagree.

3.2.1 The Utility of the Political Balance Requirement

Some opponents may imagine that is better to have the decision to prosecute made by a single prosecutor who follows the admonition of the Department Guidelines to eschew any consideration of a person’s “political association, activities, or beliefs” (U.S. Dep’t of Just., Just. Manual § 9-27.745 (2018)). We oppose this approach because we find it insufficient. It is hard for some people to rise above what are often core components of their identity.38 Besides the risk of bad-faith decision-making, we believe that good-faith attempts to stand outside of politics and make impartial decisions pose a daunting (some would say impossible) task.39

Instead, as noted earlier, we believe it is better to lean into the problem by trying to counter-balance potential political bias of the current administration’s prosecutors with the potential bias of former prosecutors nominated by other administrations. Our proposal is in keeping with the kind of checks and balances approach that motivates our Constitution’s structure.40 As discussed in the Introduction, our two-thirds concurrence requirement is intentionally borrowed from the Constitution’s impeachment requirement for Senate conviction. Just as the super-majority requirement there provides some protection against politically motivated impeachment trials, our supermajority requirement of prosecutors, with representatives of both parties, serves as a bulwark against politically motivated prosecutions.

Our proposal is also consonant with a common regulatory requirement that the members of multi-member agencies be politically balanced. Political-balance requirements (PBRs) are quite common. A majority of all multimember agencies within the federal government are subject to a political balance requirement that restricts the number of members appointed from any one political party (Feinstein & Hemel 2018). Prominent examples include the Securities and Exchange Commission, the Federal Trade Commission and the Equal Employment Opportunity Commission (Kirti & Revesz 2013, 797 (listing 27 agencies with PBR requirements)). For instance, the Federal Election Commission is composed of six members, no more than three of whom may be from the same party (52 U.S.C. § 30106 (2012)).41 Congress often requires political balance when legislators worry that the executive, through their nominations and appointments, might gain improper partisan advantage. Legislators also may suppose that there are informational advantages from having viewpoint diversity at the apex of these agencies.

Requiring political balance on a Prosecutor Jury can produce two mutually enforcing benefits. The requirement might improve deliberations by offsetting the dysfunctional tendencies of polarized groups who, inter alia, may wrongly blame others for bad outcomes (Feinstein & Hemel 2018, p. 78–81; Sunstein 2000, p. 74).42 And when deliberation fails to generate unanimity on the Prosecutor Jury, the PBR can allow substantial minorities to usefully obstruct prosecutions (Feinstein & Hemel 2018, p. 78–81 (discussing the potential obstruction rationale for PBR)).

As implemented in the agency context, PBRs are often criticized because the executive might still cherry-pick appointees from other party who nonetheless mirror the policy preferences of the President (Feinstein & Hemel 2018, p., 9).43 Our proposal is immune to this cherry-picking criticism because it calls for random selection of individuals from the opposition party. This random selection mirrors the random selection of grand jurors (Handbook for Federal Grand Jurors 2023, Handbook for Federal Grand Jurors 2023, p. 6)—except by adding on the political-balance requirement, our proposal better assures some representation by the defendant’s political peers. Moreover, in contrast to agency PBRs, our proposal can more easily identify the relevant political party, as it turns not on the party of the person being promoted (who might not be registered) but on the party of the nominating president.

3.2.2. As Compared to Special Counsels, A Prosecutor Jury Better Counters Bias

Others may suppose that a Prosecutor Jury addresses a problem that another mechanism already solves. As noted earlier, the appointment of a Special Counsel is meant to answer concerns about partisan prosecutions. But, as we pointed out earlier, history has proven otherwise. Rightly or wrongly, Special Counsels are regularly seen as partisan or biased. If they are pursuing a popular politician, that politician’s acolytes are poised to think the worst of that Special Counsel. And sometimes the staffing choices of Special Counsels leaves much to be desired. Too often, they have played into the criticisms of those they are investigating and prosecuting.

As noted earlier, one might say that the real purpose of Special Counsels is to shield the Department of Justice, in particular the Attorney General, from criticism. The Department can slough off complaints by saying “we are not in charge of the investigation and prosecution. The Special Counsel is.” Whatever happens, the Department says, it is not our fault or problem.

In contrast, our mechanism directly responds to the real problem of politicized prosecutions. Even if the prosecutor, Special Counsel or otherwise, is a pure partisan, the Prosecutor’s Jury serves as a gauntlet, a testing ground for their claims. If the prosecutor cannot secure considerable approval from professional, expert former US Attorneys, then they cannot go forward with a prosecution. And it is the presence of US Attorneys of both parties and the supermajority requirement that renders accusations of partisanship on the part of the Prosecutor Jury less plausible. That perhaps will not stop prosecutorial targets from making the farfetched claim. But it will cause many Americans to pay little heed to their charges. And that, in turn, will shore up the confidence in the judicial system.

3.2.3 Our Proposal Does Not Violate Article II

Some critics might argue that the political-balance requirement unconstitutionally impedes the President’s appointment and removal Powers. Some imagine that Congress may not restrict the President’s power to appoint and remove.44 While there has been broad presidential acquiescence in PBRs, President George H.W. Bush in 1992 noted in a signing statement that he viewed a PBR “as precatory” only (Statement on Signing the Morris K. Udall Scholarship and Excellence in National Environmental and Native American Public Policy Act of 1992, 28 Weekly Comp. Pres. Docs. 483, 507 (March 23, 1992); see alsoKrotoszynski et al. 2015, p. 976). The statement explained the President’s view that under “the Appointments Clause of the Constitution... congressional participation in such appointments may be exercised only through the Senate’s advice and consent with respect to Presidential nominees” (Statement on Signing the Morris K. Udall Scholarship and Excellence in National Environmental and Native American Public Policy Act of 1992, 28 Weekly Comp. Pres. Docs. 483, 507 (Mar. 23, 1992)). In other words, rather than imposing PBRs by law, Senators ought to insist upon balance as they give (or withhold) their consent to a presidential appointment.

Against this argument, the practice of Congress imposing qualifications for statutory offices dates back to the Washington Administration. Though a number of statutory provisions limited whom the President might nominate and appoint, Washington never objected to the laws, signing each of them (Prakash 2016, p. 175–176). Furthermore, there are decades of modern regulatory practice by dozens of agencies, which has created what some scholars have described as “the presumed constitutionality” of political balance requirements in particular (Krotoszynski et al. 2015, p. 968).45

In any event, the constitutionality of our proposal’s PBR is evident. Under our proposal, each prosecutor juror has a rather limited duty—approve (or disapprove) of the projected prosecution. Further, each juror’s involvement extends no further than making decisions as to one individual. Each attempt to prosecute a politician will require the creation of a new Prosecutor Jury. All in all, the former US Attorneys who are invited at random to serve on the Prosecutor Jury are more akin to citizens asked at random to serve on a grand jury or petit jury. Because their federal duties would be rather short-lived, Prosecutor Jurors are not officers. In U.S. v. Germaine, the Supreme Court said “the term [office] embraces the ideas of tenure, duration, emolument, and duties, and that the latter [the duties] were continuing and permanent, not occasional or temporary” (99 U.S. 508, 511 (1878)). In our view, there is no chance that the courts will regard Prosecutor Jurors as officers precisely because their duties are rather fleeting. Because Prosecutor Jurors are akin to regular jurors, the appointment and removal powers are irrelevant.

3.2.4 Our Proposal Does Not Undermine the Rule of Law

A foundational concern that critics might level at our proposal is that it undermines the rule of law. That precept, discussed earlier, may cause some to imagine that every lawbreaker must face criminal sanctions. Such a detractor might worry that by refusing to approve a prosecution, a Prosecutor Jury will, in fact, lawlessly immunize certain privileged defendants. In sum, some may suppose that making it harder to prosecute politicians is inconsistent with the rule of law.

Our response begins by noticing that few regard other procedural safeguards as undermining the rule of law. The Bill of Rights contains a host of rules related to evidence gathering, evidence admission, and prosecutions. Most Americans do not regard these protections as inconsistent with rule of law. Rather these costly rules are thought to be in service of the rule of law. We do not permit a single prosecutor to be an arbitrator of guilt, but instead require a grand jury indictment followed by a unanimous verdict from a petit jury. Furthermore, there is a judge watchfully monitoring the proceedings. Each of these rules makes it harder to secure a conviction and thus makes it possible for the guilty to escape punishment. But they nonetheless are in service of a more refined conception of the rule of law, one that does not prize the punishment of the guilty at the expense of all other values, including the protection of the innocent.

Our proposal merely adds an additional protection to those that already allow culpable people to avoid sanction. This extra hindrance is justified because there is a great harm in the perception that some popular and important public figure is suffering a politicized prosecution. When the criminal law is used to silence opposition, discredit political rivals, or simply to punish those who hold different views, the vibrancy of our democracy is at stake. As noted, we already bias the criminal process in myriad ways in favor of defendants because the harms of type I error (convicting the innocent) are deemed to be greater than the harms of type II error (acquitting the guilty). This is implicit in Blackstone’s notion that it is better that “ten guilty persons escape, than that one innocent suffer” (Blackstone 1765–1769a, p. 352). We have good reason for an even higher ratio when there is a risk of partisan prosecution, because the latter has the potential to undermine faith in our legal system in a more systemic and insidious way. By increasing confidence in prosecutions, our proposal conduces to a better, more complex, and defensible conception of the rule of law.

Singling out particular groups of people for special procedural protection also may seem inconsistent with a concern for equal treatment. But we already do just this with regard to juvenile offenders (where we believe that the stakes are different) (U.S. Dep’t of Just., Crim. Res. Manual § 1-499.121). More importantly, there are several special protections available to federal officials. As a matter of constitutional law, members of Congress have the privilege against certain arrests while attending Congress and going to, or coming from, Congress (U.S. Const. art. I, § 6, cl. 1). Legislators also have a speech and debate privilege that bars using their statements in criminal prosecutions (id.). Presidents (Nixon v. Fitzgerald, 457 U.S. 731, 748 n.27 (1982)) and judges (Barr v. Matteo, 360 U.S. 564, 570 (1959)) have official immunity that extends to civil actions. And Presidents now have immunity with respect to certain official acts. Each of these shields, either expressly in the Constitution or inferred, make it more difficult to vindicate the law in a narrow sense. But they are all in service of a more capacious sense of the rule of law.

Of greater relevance, the Justice Department already recognizes the need for greater protection for certain figures. As discussed earlier, the Department imposes special duties of consultation and approval before even investigatory steps can be taken with regard to members of Congress. And both Attorney Generals Barr and Garland have extended similar protections to cover investigations that raise election sensitivities. Our proposal stands on a similar footing. If special procedural protections for certain people are inconsistent with the rule of law, the existing rules cannot be defended. If, however, our procedural protections are conceived as part of the rule of law, and in furtherance of it, then our proposal cannot undermine law’s rule merely because it makes conviction more difficult.

Especially during these periods of extreme tribalism, there is a natural inclination to worry that one’s political opponents are so reprehensible that they will nullify the rule of law and implicitly collude to immunize guilty political defendants from prosecution. But the possibility of nullification does not cause us to turn away from the existing jury system. We believe instead that an attitude of empathy and humility is in order. If we think our political adversaries might be biased and unable to soberly judge whether or not a potential defendant is guilty, then we also must entertain the possibility that we are biased as well. Attorneys of the same political party as the potential defendant might be biased in their view of the evidence or the law, but so too might prosecutors nominated by the opposition party and serving as federal prosecutors. Both sides of the aisle may be partial. Given that only one side of the aisle controls the prosecutor’s chair, we believe it is necessary to mitigate the partiality that might infect prosecutors and to counter any perceptions of prejudice. Requiring deliberation and ultimate buy-in by some nominees of each party is a reasonable way to insulate the criminal justice system from that bias.

* * *

We believe that our Prosecutor Jury proposal is well-suited to solve (or at least mitigate) the difficulties that plague our current system. It creates a check on partisan prosecutions while still permitting many to go forward. It thereby better addresses the reality and appearance of partisan bias in decisions to prosecute, certainly as compared to a Special Counsel. Our proposal also is entirely constitutional and sound, for it does not violate the Rule of Law or any specific constitutional provisions. Finally, and most importantly, it assuages the public that a prosecution of a presidential candidate, Senator, judge, etc., is not a partisan witch-hunt. Prosecutions that occur after the filtration supplied by the Prosecutor Jury are inoculated against the claim that the Administration is pursuing its enemies using the instruments of the state. Our mechanism thus avoids corrosion and instead instills public confidence in our government.

4. SIMULATION

We thought it would be beneficial to simulate a prosecutorial jury. Here we describe our simulations and our attempts to make sense of the findings. The simulations relate to various charges against President Trump. We recognize and discuss the simulation’s numerous limitations. While the implications are limited, we think the results bolster the proposition that a Prosecutor Jury could serve a rather useful screening function.

4.1 Survey Design

As a crude simulation of how a Prosecutor Jury might produce useful information, we surveyed former US Attorneys.46 Using the Congressional Record, we first identified 554 individuals who had been confirmed by the Senate starting with Ronald Reagan’s presidency in 1981 and ending with Donald Trump’s presidency in 2020. Of these individuals, 204 had been nominated by Democratic Presidents (Clinton or Obama) and 350 had been nominated by Republican Presidents (Reagan, H.W. Bush, W. Bush, or Trump). We then searched for contact information in the form of emails and/or LinkedIn addresses.47 We were able to find contact information for 444 US Attorneys.

We sent out two waves of surveys—a February survey (which delivered at the end of the January and beginning of February 2023) and a July survey (which was delivered at the end of June and beginning of July 2023). The first wave related to the charges generated by the January 6th Committee. The second wave related to the charges brought by the New York District Attorney and by the Special Counsel in Florida. We have conducted no survey directly tied to the Special Counsel’s August 2023 D.C. indictments. Nor have we surveyed opinions on the Georgia indictments.

To these 444 individuals, we sent the following message on January 23, 2023 as the first survey, soliciting their participation with the following message:

Dear <<Name>>

We write to give you an opportunity for your voice to be heard in the weighty decision whether to indict our former President. We hope that your voice will instill confidence in the justice system.

We believe that any indictment ought to eschew political considerations and reflect the best professional judgments of prosecutors. To that end, we are asking 553 former US Attorneys to weigh in on this question in hopes of providing the Justice Department and the American people with important information about whether indicting the former president is appropriate. While your individual answer will be held in the strictest confidence, we intend to release aggregate information about the responses.

We have included here a copy of the January 6th Committee’s report that provides the basis for their referral. We recognize that the Special Counsel may have access to additional information relevant to the indictment decision. Further, we also recognize that the Special Counsel (and others) might believe that other federal charges are worthy of consideration.

To keep things tractable, we ask you to respond to questions regarding the four charges for which the January 6th Committee has issued referrals. We have added an additional question, one related to other possible charges.

To confidentially respond, click on this survey [[with hyperlink

Whether or how you respond will not be disclosed.

We will report the aggregate results to the public. We also will report the results for respondents nominated by Republican Presidents, and results for respondents nominated by Democratic Presidents. The aggregate views of former US Attorneys nominated by Republican Presidents might be particularly important in informing the American people about whether a prosecution is driven by partisanship. The Justice Department might be wary about seeking an indictment if at least a substantial minority of the Republican nominated respondents do not support seeking an indictment.

By way of background, one of us is a Democrat and the other a Republican. Like you, we hope that any decision is based on proper prosecutorial considerations rather than by animus or favor. We have taken no public stance on whether the former President ought to be indicted and do not plan on taking any such stance, except to suggest that the Special Counsel should give particular weight to your views in deciding how to proceed.

Thank you for your time and patience.

Sincerely,

Ian Ayres

Oscar M. Ruebhausen Professor

Yale Law School

Saikrishna Prakash

James Monroe Distinguished Professor

University of Virginia, School of Law

We also sent a follow-up message to non-responders on January 31, 2023, warning them that it was their “last chance” to respond and that the survey would end on February 3, 2023.48

People who clicked on the link were taken to a Qualtrics web page that, after a short-opening paragraph mandated by Yale’s human subjects committee explaining that participation was voluntary and giving contact information for people with concerns,49 asked five questions:

Should the Special Counsel seek to have a grand jury indict Donald J. Trump for Obstruction of an Official Proceeding (18 U.S.C. § 1512(c))?

Should the Special Counsel seek to have a grand jury indict Donald J. Trump for Conspiracy to Defraud the United States (18 U.S.C. § 371)?

Should the Special Counsel seek to have a grand jury indict Donald J. Trump for Conspiracy to Make a False Statement (18 U.S.C. §§ 371, 1001)?

Should the Special Counsel seek to have a grand jury indict Donald J. Trump for Inciting, Assisting or Giving Aid and Comfort to an Insurrection (18 U.S.C. § 2383)?

Should the Special Counsel seek to have a grand jury indict Donald J. Trump on any other charges?

If yes, which charge or charges

The survey ended by asking the name of President who nominated them to be a United States Attorney and by providing an open text box to allow respondents to provide any comments or feedback. The solicitation messages included subject-specific Qualtrics hyperlinks which allowed us to identify the US Attorney responding.50 The subject-specific links allowed us to judge whether subjects misreported their nominating President or responded multiple times.

Finally, our subjects were randomly assigned to one of two groups. Subjects who were randomly assigned to the “Less Forced” group were given three possible answers for each of the five questions: “Yes,” “No,” and “Unsure” (Web App., 2–3). Subjects who were assigned to the “More Forced” group were presented with just two possible responses: “Yes” and “No” (Web App., 4–5). The online survey was programmed to permit subjects in either group to leave questions unanswered—so that subjects in either group could decline to answer particular questions and subjects could explain in the feedback box their reasons for leaving some or all questions blank. Our goal in adding this randomized component was to be able to make some causal inferences about the strength of conviction or belief. For example, if we learned that 80% of the subjects in the “More Forced” group responded that Trump should not be indicted for Obstructing an Official Proceeding, but that subjects in the “Less Forced” Group only showed a 30% support for not seeking an indictment (with 50% of respondents opting for “Unsure”), the shift in response would suggest that a substantial proportion of the 80% had some doubts about whether the charge was appropriate.51

At the end of June and beginning of July 2023, we sent a second wave of surveys to the same US attorneys. This time we asked them about their opinion on the allegations contained in the Manhattan District Attorney’s Statement of Facts and the Special Counsel’s Southern District of Florida Indictment, respectively. The following message was sent:

Dear <<Name>>

We write again to give you an opportunity for your voice to be heard regarding the recent decisions to indict our former President.

As we said in our original request:

We believe that any indictments ought to eschew political considerations and reflect the best professional judgments of prosecutors. To that end, we are asking 553 former US Attorneys to weigh in on this question in hopes of providing the Justice Department and the American people with important information about whether indicting the former president is appropriate. While your individual answer will be held in the strictest confidence, we intend to release aggregate information about the responses.

Our initial request focused on the four charges referred to the Justice Department in the January 6th Committee’s report.

Today, we are asking whether you believe the two indictments, one state and one federal, were appropriate. In particular, we seek your professional judgment on the allegations contained in (1) the Manhattan District Attorney’s Statement of Facts and (2) the Special Counsel’s Southern District of Florida Indictment.

We recognize that the prosecutors may have additional information relevant to their decisions to indict.

To confidentially respond, click on this 2-question survey

We will not disclose whether or how you respond.

We will report the aggregate results to the public. We also will report the results for respondents nominated by Republican Presidents, and results for respondents nominated by Democratic Presidents. The aggregate views of former US Attorneys nominated by Republican Presidents might be particularly important in informing the American people about whether a prosecution is driven by partisanship.

By way of background, one of us is a Democrat and the other a Republican. Like you, we hope that any decisions were, and will continue to be, based on proper prosecutorial considerations rather than by animus or favor. We have taken no public stance on whether the former President ought to be indicted or whether the two indictments were appropriate. We do not plan on taking any such stance.

Thank you for your time and patience.

Sincerely,

Ian Ayres

Oscar M. Ruebhausen Professor

Yale Law School

Saikrishna Prakash

James Monroe Distinguished Professor

University of Virginia, School of Law

Upon clicking the link, the respondents were taken to another Qualtrics web page that asked three questions:

Do the allegations contained in the Manhattan District Attorney’s Statement of Facts convince you that an indictment was appropriate?

Do the allegations contained in the Special Counsel’s Southern District of Florida Indictment convince you that an indictment was appropriate?

Please provide any additional comments. We appreciate your feedback.

4.2 Response Rate and Caveats

Table 1 displays the number of US Attorneys that we solicited and the number that responded:

As shown in the table, for the February survey, the overall response rate of those whom we solicited was just 11.7 percent, with even lower response rates of Republican nominated attorneys and of Trump nominated attorneys of 9.8 percent and 5.9 percent, respectively.52 The July survey responses rate of 11.3 percent was slightly lower than that of February. There was also a drop in the response rate of the Democratic nominated attorneys to 12.6 percent. The response rate of Republican nominated attorneys increased slightly to 10.5 percent but there was a notable drop in the response rate to 2.9 percent among Trump nominated attorneys.

First, these response rates are extremely low and limit the useful inferences that can be drawn from the survey responses. Respondents may not share views that are representative of the larger group of former US Attorneys who did not respond. Relatedly, the US Attorneys for whom we were able to find contact information might have different views than the Attorneys for whom we were unable to find such information.

Second, the views of respondents might change were they asked by the Attorney General to participate in a formal Justice Department process, rather than by two law professors. Some US Attorneys commented that it would be inappropriate for them to advise the Justice Department. Some of these comments emphasized their limited access to evidence—even though our messages directed their attention to the 845-page report of the January 6th Committee which was replete with sworn witness testimony (H.R. Rep. No. 117-663 (2022)). For example, one former the US Attorney wrote:

Thanks for reaching out, but I think it would be inappropriate for me to tell DOJ what to do regarding a charging decision. They are privy to facts that are completely unknown to me, or other members of the public. While there certainly is a lot of evidence to suggest criminal activity, DOJ is best situated to make that assessment.

Another observed:

I would make a prosecutorial judgment if I had access to all of the evidence—which I do not. Short of that, I don’t see how this endeavor—however well intended—is all that helpful.

Others more explicitly called upon us to abandon the project.

This survey and your plan to publish the results is inappropriate and damaging to our system of justice. A prosecutor is responsible for making an independent, fact-based, evaluation whether to seek an indictment. He or she does not poll other prosecutors to decide whether to make a charging decision. To do so would be improper. Your survey asking former prosecutors to weigh in on this matter asks people who did not hear the evidence, did not evaluated the credibility of the witness, did not take time to study the record developed, is at least unrealistic if not silly and naive. That you would publish any of your “results” may well bias members of the public in evaluating and trusting the decision reached. What you are doing is reckless and I encourage you and your colleague to abandon it. You have no right to suggest to the special counsel that your survey has any valuable [sic] or credibility in deciding the proper judgment in this matter. That is a decision that is entrusted to the special counsel alone.

Another respondent put it more bluntly:

This is outrageous! No one can answer this question without reviewing the evidence. Any prosecutor will tell you that. Shame on you.

Other subjects, in declining to participate, focused not on their access to evidence but on whether it was appropriate for the Department of Justice to receive survey advice from former prosecutors. For example, one attorney opined:

What I really think is that the Attorney General should stop appointing special counsel and should make the God damn call.

Another respondent, in addition to mentioning evidentiary access, went on to question whether they were capable of rising above their own potential political bias:

Thank you for offering to include my thoughts on this issue. I do, however, decline to participate and would like to explain why. I appreciate the way in which you have communicated this and how you have set up the survey. I do believe that you are trying to put together an honest survey. But while the committee’s report provides a lot of information, it is not the same record that a federal prosecutor using the tools available in a grand jury investigation would have. There may be additional information that a prosecutor would have and would consider that could affect the charging decision. I don’t believe that I’m in a position to say what decision would be made if I don’t see what the prosecutor sees. Also, an exercise like this, no matter how well-intentioned, is inevitably affected by the political views of the respondents. It is much easier for all of us who have left the prosecutor’s chair to be affected by our views. To be clear, while I am a Republican, I am not and never have been a Trump supporter and would like to see him no longer in public life. I think it is bad for my party and my country. But in coming to my decision as to what I think should happen, can I put those views aside? Do I lean over backwards to avoid that opinion and go the other way? When in office, I recused myself from a case in which I believed I could not view the matter impartially and that I would do neither side justice if I stayed in. In addition, whatever the results of this survey, it could serve to affect the decision and I believe the decision should not be driven by concerns over how a group of people who have not seen all of the evidence might view the prosecutorial decision.

How best to respond to these concerns and criticisms? As noted, we admit that only limited inferences can be drawn from our survey—both because our response rate is low and because the views expressed by individuals formally charged with advising the Justice might be different than the views expressed in our survey. We further agree that the evidence presented to a Prosecutor Jury likely would be different than the publicly available information contained in the Jan. 6th report. That said, we imagine that the publicly available testimony and other evidence might have been sufficient for some respondents to form views as to the advisability of seeking some indictments. Or respondents could have expressed a view conditional on there not being additional (exculpatory or inculpatory) evidence.

The separate concern that it is inappropriate for the Justice Department to seek outside advice would be obviated (or blunted) by our proposal for the Justice Department to empanel a Prosecutor Jury. While a single individual—the Attorney General or a Special Counsel—might have to make the ultimate decision whether to prosecute, that decider already benefits from internal advice in the form of consultation with the Public Integrity Section (as discussed above). And the existing system already requires the consent of a grand jury before a prosecution can proceed. Former prosecutors serving on Prosecutor Jury would be providing a screening function analogous to that of the grand jury, as we argued above, in order to create a reasonable counterweight to the threat of a partisan prosecution. We agree that it is difficult to put aside personal political commitments. But, as we noted earlier, that concern applies to prosecutors no less to jurors on the Prosecution Jury. Demanding some assent from prosecutors of the same party is a reasonable way to reduce the risk of improper partisan influence on the part of the prosecution—just as the supermajority Senate requirement for conviction helps insulate the impeachment process.53

4.3 Analysis of Survey Responses

Our core results are summarized in the following table showing the proportion of respondents that supported an indictment for particular offenses.

As might be expected, Table 2 shows a partisan differential. Respondents nominated by Democrats were much more likely than respondents nominated by Republicans to answer “yes” to our questions regarding whether Special Counsel should seek indictments against the former President. Indeed, in the February survey, the Democratic proportions are two to three times larger than the Republican proportions and, except for the question concerning additional potential charges, all these differences are statistically significant (p < 0.05).54 For the July survey we found that the Democratic proportions convinced by the allegations in the New York State Indictment was also almost three times as that of the convinced Republican nominated attorneys (p < 0.05).

To our mind, the most important result in the table is bipartisan support for the Florida indictment. Not only did 85 percent of the Democratic-nominated attorneys say they were convinced but also a substantial majority (63.3 percent) of the Republican-nominated attorneys said they were convinced that the indictment was appropriate. The bottom row of Table 2 labeled “Balanced Average” is the simple average of the two reported percentages of the two nominee groups. Giving equal weight to the two nominee groups in this fashion, we find that only the Florida indictment related to classified documents meets the two-thirds (66 percent) Prosecutor Jury Standard, with almost three-quarters (74.2 percent) as the balanced (bipartisan) average.

Second, all the other charges fell substantially short of our two-thirds requirement. To achieve two-thirds support in our balanced Prosecutor Jury, the DOJ would need to convince at least one-third of Republican nominees to back an indictment, and this percentage would be sufficient only if 100 percent of Democratic nominees supported indictment. But as Table 2 shows, the largest support for the other charges among Democratic nominees falls twenty five percentage points below 100 percent and the largest support of Republican nominees falls more than four percentage points below the 1/3. In fact, none of the other charges received even a majority of balanced support.

Third, consider the February survey results in conjunction with the Washington, D.C. indictment. The D.C. charges filed on August 1, 2023 included three of the referred charges mentioned in the table: obstruction of an official proceeding, conspiracy to defraud the United States, conspiracy to make a false statement.55 The survey results might suggest both an error of inclusion and an error of omission in the Special Counsel’s charging decisions. In other words, Jack Smith arguably brought one ill-advised charge and left out one more promising one. The Special Counsel’s decision to include the charge of conspiracy to defraud the United States might have been unwise as it received the least support across all charges among both the Democratic nominated and Republican nominated attorneys. Consistent with our survey results, the Wall Street Journal regarded this charge as particularly problematic (Editorial Board 2023b (“Special counsel Jack Smith’s broad theory of fraud has dangerous implications.”)). Second, the decision to not charge inciting, assisting, or giving aid and comfort to an insurrection (18 U.S.C. § 2383) might have been an unwise omission because, of the January 6th charges, that charge received the greatest support (75 percent) among the Democratic nominated attorneys and had the highest balanced average. In other words, our survey results suggest that this was the strongest referred charge and yet the Counsel seems not to have brought it before the grand jury.

Fourth, reflect on the ability of respondents to transcend political tribalism. Unlike the lineup of Justices in the politically charged issues in Bush v. Gore (531 U.S. 98 (2000)), some of the Republican-nominated US Attorney’s transcended crude party solidarity by indicating that indictments should be pursued against our former Republican president. Likewise, some Democratic-nominated U.S. Attorneys indicated that the Republican President should not be indicted.56 This is heartening.

Fifth, consider the nuanced results across the four charges referred by the January 6th Committee. As noted earlier, the Conspiracy to Defraud the United States charge received the least support–with the lowest proportion of support from Democratic nominees (54.2 percent) and the lowest proportion of support from Republican nominees (21.4 percent).57 In contrast, Inciting, Assisting or Giving Aid and Comfort to an Insurrection received the highest support from Democratic-nominees (75.0 percent) and Obstruction of an Official Proceeding received the highest support from Republican nominees (28.6 percent).58

Finally, looking across the two portions of the July survey we find still more nuance. Respondents to the second wave showed balanced support for the Florida indictment that outstripped the support of the New York charges by more than 25 percentage points. In other words, the same pool of respondents reacted rather differently to the two indictments. Moreover, if we compare the responses to the Florida indictment to all the other survey questions, we see that the percentage who endorsed the Florida indictment greatly exceeds—by more than 20 percentage points—the support for any other indictment.

The other results are also somewhat illuminating. The last (fifth) question on the February survey asked whether the Special Counsel should seek indictments on any other charges. Only four respondents (three Democratic-nominated and one Republican-nominated) responded affirmatively to this question, and each suggested different charges. The four recommended charges were:

“Refusal to return classified documents.”

“Failure to carry out the duties/oath of office to defend the constitution. Dereliction of duties to protect the safety of people at U. S. Capitol on 6 Jan 21 and U. S. citizens in general.”

“They should investigate whether Trump should be indicted for violating the Foreign Corrupt Practices Act. It is common knowledge that foreign officials frequently rented space in Trump’s hotel at exhorbitant [sic] rates, sometimes without using the space or giving it to groups friendly to Trump.”

“Felony murder”

The responses to the fifth question uncover rather little support for any additional charges. At first, we were surprised by the felony murder comment and worried that it might suggest a flippant or ill-considered response. But this response from a Republican-nominee (who also supported indictment on each of the four referred responses) may have a certain coherence. If one believes that President Trump committed a felony by inciting or assisting an insurrection, then one might conclude that he is culpable for those that died in connection with that violence.59

We also looked within the February and July data (i) to indirectly assess the views of those who did not respond, and (ii) to gauge the strength of conviction of our responders. First, we analyzed whether the views of responders to our follow-on solicitation differed from the views of responders to the initial solicitation. The follow-on responders who showed initial reluctance to participate may have views that are more representative of the United States Attorneys whose reluctance caused them never to respond.

Table 3 shows the results of our analysis. For the four charges put forth by the January 6th committee, one sees that the proportion of respondents supporting indictment falls in the follow-up group relative to the proportion shown in the initial group. For example, 83% of the initial responses from Democratic nominees supported indictment on the Obstruction of an Official Proceeding charge, but this proportion fell to 58.3 percent in the follow-up responses from Democratic nominees.60 And similarly, the proportion of Republican nominees supporting indictment on this charge fell from 30.8 percent among the initial responders to 26.7 percent among the follow-up responders. The reduced support across the board for indictment in the follow-up group is another piece of evidence weighing against the indictment of President Trump, at least as to the January 6th related charges. One might speculate that as compared to respondents, non-respondents were more likely to oppose indicting the former President.

In contrast, Table 3 shows that for the July survey both the New York State Indictment and the Florida Indictment, the proportions of Democratic and Republican nominees in support of the indictment increased in the follow-up responses. In particular, the difference between the initial and follow-up proportions of Republican nominees who are in support of the indictment in New York which increased from 13.6 percent to 50 percent and is statistically significant (p < 0.05). This result might suggest that support for these indictments is even stronger among non-respondents.

Finally, we analyzed, for the February survey, the impact of assigning respondents to the “Less Forced” condition (in which they were presented with “Yes,” “No,” or “Unsure” options) relative to assigning respondents to the “More Forced” condition (in which they were present with just “Yes” or “No” options). Table 4 shows the results of our analysis.

The first row of Table 4 shows the percentage point difference for Democratic nominees who responded to the Obstruction of an Official Proceeding charge depending on whether the nominee was randomly assigned to the “Less Forced” or “More Forced” condition. Democratic nominees in the “Less Forced” condition were 21.7 percentage points more likely to decline to answer or to respond that they were unsure. Interestingly, this shift in response came disproportionately from Democratic nominees having a substantially lower likelihood of responding “Yes” in the “Less Forced” condition—dropping 20.3 percentage points from 81.8 percent in the “More Forced” condition to 61.5 percent in the “Less Forced” condition. Because this change is the treatment effect of a between group randomized experiment, the statistically significant result can be interpreted causally. Adding the “Unsure” option caused Democratic nominees to reduce their likelihood of signaling that Donald Trump ought to be indicted. This result suggests that a substantial proportion of respondents supporting Donald Trump’s indictment on this first charge harbor some degree of uncertainty in their conviction. In contrast, the percentage of Democratic nominees who expressed the view that the former President should not be indicted on this charge hardly changed even when they had an express option to express their uncertainty.

The second row of the table estimates the results with regard to our twenty eight Republican-nominee respondents. However, instead of observing an expected positive shift toward Unsure\Declining to Answer as we move from the “More Forced” to the “Less Forced,” we surprisingly see a statistically significant decrease in the proportion of Unsure\Declining to answer, and increases in both the proportion of Republican nominees who answered in the affirmative and in the negative. While this shift is inconsistent with certain forms of rational choice theory,61 it might be produced by certain cognitive distortions or be a byproduct of different types of subjects being randomly assigned to the two different groups.62 And while Table 4 merely reports results regarding the “Obstruction of an Official Proceeding Charge,” we found analogous results with regard to the other three referred charges (Web App., p. 5–6).

Step back for a moment.63 We emphasize again the limited inferences that can be drawn from our results, both because the narrow set of responders might have non-representative views and because the views of respondents might change if they participated in a formal, DOJ-sponsored Prosecutor Jury. But with these important caveats, our survey results did not make out strong support for the Special Counsel’s January 6th indictment. We see four reasons:

• (i) The politically-balanced averages do not show a strict majority supporting indictment (Table 2)—which is far short of our proposed two-thirds requirement;

• (ii) Only 21 to 28 percent of the former President’s own party support indictment (Table 2)—which is short of our implicit one-third requirement;

• (iii) The proportion of indictment support declined significantly in the follow-up responses (Table 3)—consistent with possibility that non-responders might have less support for indictment; and,

• (iv) The proportion of indictment support among Democratic nominees declined significantly in the “Less Forced” condition (Table 4; Web App., p. 5–6)—suggesting less strength in their conviction/certainty that seeking an indictment would be appropriate.

And yet, the July survey suggests substantial support for the Florida federal indictment—with substantial majorities of both Republican-nominated and Democratic-nominated attorneys (with even more support among the follow-on responders) (Table 2).

We conducted the survey primarily as a proof-of-concept exercise. Even with our limited ability to contact former US attorneys, to convince them to participate, to present relevant evidence in a context where they could deliberate, we have at a minimum shown that it is possible to elicit information that varies by charge, has some indicia of integrity, and where response transcended crude party allegiance. Each of these is a welcome result.

5. CONCLUSION

In an era of political tribalism, suffused with mistrust of the other party, admonitions to prosecutors to eschew consideration of “political association, activities, or beliefs” are more necessary than ever (U.S. Dep’t of Just., Just. Manual § 9-27.230 (2018)). In a nod towards the need for further safeguards, the Department of Justice created additional precautionary rules for members of Congress and candidates for federal office. These instructions require consultation with higher-ups, a process that we favor. Care and deliberation can be an antidote to the improper consideration of politics.

But, as should be obvious, Attorneys General and other high DOJ officials can be biased as well. And, just as importantly, members of the public can legitimately perceive such supervision and decisions as partial. Hence while consultation rules will stop some biased investigations and prosecutions, they certainly will not halt them all. Moreover, even if consultation somehow put a stop to all biased investigations, there still will be sincere and legitimate fears that the cases that proceed to a prosecution reflect a conscious or unconscious prejudice against a partisan enemy.

The existing rules are insufficient to meet the needs of our tribal moment. If prosecutors are seen as biased and if the resulting prosecutions are viewed as tainted, or even illegitimate, there is a need for a reform to counter that perception. Like the juries required by the Constitution, our Prosecutor Jury is designed to remedy a structural problem, one of reality and perception. The Prosecutor Jury, since it is composed of experienced professionals, can render expert judgments about whether charges are appropriate. The Prosecutor Jury, because it enjoys political balance and requires significant support from individuals affiliated with both parties, can better fend off charges of partisanship and bias. Any charges that emerge from such a Jury seem relatively insulated from the fear that the charges merely reflect party prejudices.

We recognize, of course, that this reform comes at a price. There will be fewer charges brought against politicians because some prosecutors will not seek to bring charges at all. After all, some prosecutors may fear that the Prosecutor Jury would rebuff them. And when prosecutors do bring charges to the Prosecutor Jury, the latter may reject some or all of the proffered charges.

This is a price worth paying. To begin with, the failure to bring some charges may not be a cost, insofar as the petit jury may reject the charges down the road. If a Prosecutor Jury rejects charges, it may merely save later effort and labor. Furthermore, if the public’s confidence in the system materially increases, that is to be celebrated. Democrats would know that Republican prosecutors cannot easily target and harass their favorite politicians. And Republicans would have the same confidence. In our estimation, the benefits outstrip the costs, making our reform a positive addition to the ideal of the rule of law.

The bleak alternative is the present path. America could soon devolve into a war of one party against the other. Weaponized prosecutions by one party would be followed by weaponized prosecutions by the other, with a descent into a banana republic. We do not predict this outcome. But with earlier cries of “Lock her up” (Gass 2016) followed by at least some cries of “Lock him up” (Becker 2023) we believe that now is the time to act to forestall this rhetoric from turning into reality.

Works Cited

1

The New York Times referred to the Manhattan District Attorney’s legal theory as “largely untested” (Rashbaum et al. 2023).

2

U.S. Const. art. I, § 3, cl. 6 (“And no Person shall be convicted without the Concurrence of two thirds of the Members present.”).

3

U.S. Const. art. I, § 5, cl. 2 (“Each House may... with the Concurrence of two thirds, expel a Member.”).

4

Of course, we do not suppose that non-politicized House impeachments will necessarily succeed in the Senate. Those with the purest of motives can be seen as partisan hacks.

For an argument that the two-thirds conviction threshold means that impeachments require some bipartisanship, see Amar (2016, 93–94, 280–281, 290–291, 308–310).

President Trump’s lawyers have argued that a failure to convict in the Senate prevents future prosecution of analogous offenses (Liptak 2024). There is a certain similarity between Trump’s claim and our proposal, in the sense that bipartisanship is necessary for a Senate conviction and for a Prosecutor Jury. Having said that, our proposal explicitly imposes a check on prosecutions. There is no room for doubt. In contrast, the Impeachment Judgment Clause of Article I, section 3, does not expressly provide that if the Senate fails to convict an officer, a future prosecution grounded on the same offenses is prohibited. Further, a Senate failure to convict may not reflect innocence but may instead reflect a lack of jurisdiction, including that the offenses are not impeachable offenses and that person tried is no longer subject to impeachment.

5

A former US Attorney pointed out that some grand juries unearth facts not previously known by prosecutors. Hence a prosecutor knows more about the merits after a grand jury’s proceedings. Inserting the Prosecutor Jury at this stage, post indictment, allows the Prosecutor Jury to benefit from this extra information.

6

See also Kaminski et al. (2009, p. 322), who states that “he is not exempt from a trial, if he should be guilty, or supposed guilty, of [treason] or any other offence,” and Prakash (2021, p. 62–84), who demonstrates that the Constitution’s text, history, structure, and practice reveal that presidents do not enjoy constitutional immunity from criminal prosecution and punishment.

7

In Nixon v. Fitzgerald (457 U.S. 731, 798 (1982)), Justice Harry Blackmun writing in dissent memorably characterized as “the foundation of our national jurisprudence” the principle that “that no man, not even the president of the United States, is absolutely and fully above the law” (see alsoKalt 1996, 808).

8

As noted earlier, Trump v. United States does not bar the prosecution of former presidents. Instead, the case makes it difficult, and in some cases impossible, to prosecute a former president for his official acts. Yet former presidents remain subject to criminal prosecutions for their non-official acts. Further, they have a legal duty to comply with the law, whether or not they can be prosecuted for their official acts that might violate the law.

9

Of course, the President may recess appoint a U.S. Attorney.

10

Prosecutions that target individuals for their political views or affiliation also offend the First Amendment Heffernan v. City of Paterson, 578 U.S. 266, 273 (2016) (“When an employer demotes an employee out of a desire to prevent the employee from engaging in political activity that the First Amendment protects, the employee is entitled to challenge that unlawful action under the First Amendment.”); Hartman v. Moore, 547 U.S. 250, 256 (2006) (holding that “the law is settled that as a general matter the First Amendment prohibits government officials from subjecting an individual to retaliatory actions, including criminal prosecutions, for speaking out”); see alsoNewland & Parker (2020) (“A prosecution brought in retaliation for the subject’s political expression clearly violates the First Amendment.”).

11

The Justice Manual also reenforces these substantive duties against political favoritism by imposing whistleblowing duties on all Department employees: “Department employees shall report to their supervisor” any “evidence or non-frivolous allegation that” a Department attorney or law enforcement officer “engaged in professional misconduct” (see alsoNewland & Parker 2020 (“Department personnel tasked with working on an unlawful investigation or prosecution should, at a minimum, notify their supervisors, the Office of Professional Responsibility, and the Office of Inspector General, and – if legally permissible consistent with their obligations to protect government privileges – Congress.”)).

12

The Ethics in Government Act of 1978 (5 U.S.C. §§ 101–505) mandates that the Attorney General report annually to Congress on the activities of the Public Integrity Section. The Act was passed in the aftermath of the “Saturday Night Massacre,” where President Nixon ordered Attorney General Elliot Richardson to fire Special Prosecutor Archibald Cox, which prompted Richardson and his deputy, William Ruckelshaus to refuse to comply with the President’s order and to resign their position (O’Keefe & Safirstein 1982, p. 117–118).

13

Shortell (2020) notes that “[t]he Department has a strong interest in the prosecution of election-related crimes, including those involving corruption of the election process. Yet we must investigate and prosecute those matters with sensitivity and care to ensure that the Department’s actions do not unnecessarily advantage or disadvantage any candidate or political party.”

15

U.S. Const. art. VI (“[A]ll executive and judicial Officers,... shall be bound by Oath or Affirmation, to support this Constitution....”).

16

Id. art. III, § 3, cl. 1 (Treason Clause); 18 U.S.C. § 2381 (Federal Crime of Treason).

17

28 U.S.C. § 544 (“Each United States attorney, assistant United States attorney, and attorney appointed under section 543 of this title, before taking office, shall take an oath to execute faithfully his duties.”).

18

See 18 U.S.C. § 1503 (criminalizing the act of influencing or injuring officers or jurors in United States courts).

19

Cf. Berger v. USA, 295 U.S. 78, 88 (1935) (“[The prosecutor] may prosecute with earnestness and vigor—indeed, he should do so. But, while he may strike hard blows, he is not at liberty to strike foul ones. It is as much his duty to refrain from improper methods calculated to produce a wrongful conviction as it is to use every legitimate means to bring about a just one.”).

20

US v. Nixon, 418 U.S. 683, 693 (1974) (“[T]he Executive Branch has exclusive authority and absolute discretion to decide whether to prosecute a case.”).

21

For example, President Gerald R. Ford granted “a full, free, and absolute pardon unto Richard Nixon for all offenses against the United States which he, Richard Nixon, ha[d] committed or may have committed or taken part in during the period from January 20, 1969 through August 9, 1974” (Proclamation No. 4311, in 88 Stat. 2502 (1974)).

22

Tomco (2020), for example, notes that “[b]etween 2020 and the first week of 2021, the former president used the term RINO 32 times in his Twitter posts. The label was aimed at state election officials in Philadelphia; governors in Georgia, Massachusetts, Maryland and Arizona; and congressmen who accepted the results of the 2020 presidential election.”

23

Ackerman (2012) proposes a Popular Sovereignty Initiative to allow amendments if approved by at least 60% of Americans in two successive presidential elections.

24

The appearance of partisanship is susceptible to empirical testing. For example, follow-on research could inform a randomly-selected treatment group about the results of our survey of former prosecutors and then test whether the treatment alters views on the legitimacy of the Trump prosecutions relative to an uninformed control group.

25

If there is a concern that some former US attorneys are too old to participate, one might limit invitations to the last twenty years or the last three Presidential administrations of each political party.

26

Prosecutions that were vetted and approved, after deliberation, by at least one-third of randomly chosen former Republican-nominated US attorneys might instill more confidence.

27

For a discussion of these recusal mechanisms, many of which are informal, see Green & Roiphe (2017, p. 491–499).

28

We would not release the actual Juror votes on individual charges. We believe that doing so might exacerbate the sense that Jurors are themselves politicized. However, as noted, when the Prosecutor Jury recommends a prosecution, we would have the government disclose all the presented charges, including those which failed to secure a supermajority.

29

For example, see U.S. Dep’t of Just., Just. Manual § 9-27.220 (2023), which provides guidance for indictment decisions.

30

U.S. Const. amend. V (“No person shall be held to answer for a capital, or otherwise infamous crime, unless on a presentment or indictment of a Grand Jury, except in cases arising in the land or naval forces, or in the Militia, when in actual service in time of War or public danger[.]”).

31

One possibility is that a federal statute might shield elected state officials only in those states that pass analogous Prosecutor Jury laws to cover federal officials. Such a provision in federal law might give state legislators more of an incentive to enact a Prosecutor Jury provision so as to gain more protection from any future federal prosecution.

32

However, federal law might explicitly empower the Department to empanel a new Prosecutor Jury and present it with additional evidence.

33

The District Judge ordered a mistrial after discovering the worst case of prosecutorial abuse he had ever seen. The Attorney General, Eric Holder, subsequently moved to have the case dismissed (Bolstad & Mauer 2016).

34

We exclude declared candidates for federal or state elective office because to include them might create too large a chance that individuals with little interest or likelihood of election will strategically declare their candidacy just to gain the additional procedural protection. Hence while some candidates for office will benefit from Prosecutor Jury protection, because they fit into other categories (e.g., they hold office or used to hold office), “outsider” candidates will not.

35

Despite having worked on the campaigns of numerous politicians, Stone has never held an elected federal or state office. Neither has he received more than 10% of vote in a general federal election or general statewide office, nor ever held a Senate-confirmed office (Roger Stone: Trump Ally, Political Strategist, Nixon Fan and Russia Probe Defendant 2020).

36

Our proposal might lead some presidents to appoint fewer professional prosecutors who would then be expected to show more party loyalty if called upon to sit as jurors. One could imagine a version of our proposal where a super-majority of Senators would have to approve before a prosecution could proceed. Such a system would be somewhat analogous to a position President Trump’s attorneys have argued with regard to immunity—that a former president has absolute immunity from criminal prosecution for actions performed within the outer perimeter of their official responsibility while serving as President, “so long as he was not both impeached and convicted for those actions” (United States v. Trump, No. CR 23-257 (TSC), 2023 WL 8359833, at *3 (D.D.C. Dec. 1, 2023), aff’d, 91 F.4th 1173 (D.C. Cir. 2024), vacated sub nom., Trump v. United States, No. 23-939, 2024 WL 3237603 (U.S. July 1, 2024)).

37

But as part of our all-things-considered test, we believe that there should be consideration of whether a prosecution is in the national interest given the likelihood of jury nullification. A likelihood of ultimate acquittal might lead a prosecutor to consider whether an adequate non-criminal alternative to prosecution exists. For example, if the Schultz-Raskin Bill (H.R. 7906, 117th Cong. (2022)), which creates “a civil action for disqualification under section 3 of the 14th Amendment to the Constitution” and empowers the Attorney General to initiate such action, became law, Prosecutor Jurors might conclude that the federal interest was adequately served by this non-criminal alternative.

38

One might consider structural alternatives to shore up prosecutors’ political independence, for example, by barring them from future elected or appointed office, or by adopting sustained and separate training tracts. See Simone Benvenuti, The European Judicial Training Network and Its Role in the Strategy for the Europeanization of National Judges, 7 Intl. J. Ct. Admin. 59 (2015).

39

40

As Madison wrote in Federalist 51, “Ambition must be made to counteract ambition” (Shapiro 2009, p. 264). The search for philosopher statesmen is akin to the search for “angels” to govern (id. (“If men were angels, no government would be necessary.”)).

41

One of the earliest uses of a PBR arguably reflected a congressional desire to reduce the likelihood of improper political favoritism. In 1882, the Pendleton Act authorized the President to “appoint, by and with the advice and consent of the Senate, three persons, not more than two of whom shall be adherents of the same party, as Civil Service Commissioners” (Pub. L. No. 47-27, 22 Stat. 403 (1883); see alsoKrotoszynski, Jr. 2015, 966). Given that a central purpose of the Act was to eliminate political patronage, it was natural to require representation from multiple parties in the rulemaking process. The Pendleton Act used a PBR as a means to reduce ability of the executive to deploy politically-motivated carrots; our proposal uses a PBR to reduce the executive’s ability to deploy politically-motivated criminal sticks.

42

Sunstein (2000, p. 74) defines group polarization as involving “members of a deliberating group predictably mov[ing] toward a more extreme point in the direction indicated by the members’ predeliberation tendencies.” Feinstein & Hemel (2018, p. 78–81) discuss the potential deliberative advantage of PBR.

43

“[P]artisan balance requirements had at most a modest impact on the ideological composition of multimember agencies from the late 1970s to the early 1990s but a stronger effect from the mid-1990s onward” (see alsoHo 2007, p. 3–4).

44

For example, Calabresi & Prakash (1994) argue that the Constitution and historical evidence establishes that the President is empowered to administer all federal laws, and Prakash (2006) argues that the President has a constitutional power to remove.

45

See also Shurtleff v. US (189 U.S. 311, 313 (1903)), which, after quoting statute requiring political balance in appointments to Board of General Appraisers, found “[t]here is, of course, no doubt of the power of Congress to create such an office as is provided for in the above section,” and Myers v. United States (272 U.S. 52, 264 (1926) (Brandeis, J., dissenting)), which listed various restrictions narrowing the President’s choice of nominees, including “restrictions on the power of nomination by requiring political representation” that have existed in “a multitude of laws” and “[have] been exercised by Congress continuously since the foundation of the Government.”

46

The Yale Institutional Review Board (human subjects) approved our surveys.

47

Details of our method can be found in our web appendix (Web App., p. 1).

48

A copy of the follow-up solicitation is included in the web appendix (Web App., p. 1).

49

A copy of the full survey, including this paragraph, can be found in the web appendix (Web App., p. 2–5).

50

We cannot be sure that only former US Attorneys responded. For instance, if someone else had access to the subject’s email account, that other person could have clicked on the link and responded. In searching for contact emails, we rejected accounts that had indicia of group access. For example, we did not solicit responses by sending emails to a general law firm account.

51

Similar randomization toward more and less forced responses has been used in other studies. For instance, see Ayres & Tang (2020).

52

We had hopes of breaking out the aggregate views of US Attorneys who had been nominated by President Trump, but cannot, because we precommitted to only disseminating aggregates of at least ten individuals.

53

The “creative themes” that we chose to survey were just those that had been referred to the Justice Department by the January 6th committee. Far from endorsing these theories of criminal liability, we merely sought to create a space to empower experienced prosecutors to give voice to their views about these weighty issues. It is a sign of our tribalized times that our attempt to create a modest bulwark against politicized prosecutions can itself be regarded as political.

54

While the responses of Democratic and Republican nominees were statistically different in our sample, one way to underscore the inferential limitations of our low response rate is by retesting for statistical significance after assuming that all non-respondent answers would have supported the null hypothesis of no difference. This can be done by assuming that both groups of Republican and Democratic nominee non-respondents would have reported the same proportion of support as the pooled mean observed in our sample. When we attribute these responses to non-responding former US attorneys, the party differences in our sample converge on the pooled mean. And even though we have a larger implied dataset (of 444 instead of 52) to identify statistical disparities, none of the party differences depicted in Table 2 remain statistically significant when reanalyzed on this non-response-diluted basis.

55

The August 2023 indictment also included a fourth charge, the deprivation of rights under the color of law, which was not charge that the January 6th Committee referred. We did not conduct a survey on this charge (Tracking the Trump Criminal Cases 2023).

56

We also were able to assess other dimensions of respondent integrity by testing to see whether respondents any respondents attempted to “stuff the ballot box” by responding more than once or misreporting the President who nominated them. Especially because our solicitation letter explained that the views of Republican-nominated appointees might be particularly important, one might imagine that Democratic-nominated appointees who believe that President Trump should be indicted would have a strategic reason to say that they were appointed by a Republican president. In analyzing the respondent-specific hyperlinks, we uncovered no evidence of either multiple responses or misreporting of nominating president. Another measure of response quality is the length of time that respondents took to complete the survey. The average respondent answering the Qualtrics survey took over 9 minutes, although that included one respondent who took just 9 seconds to report “Unsure” to all five questions.

57

Using these proportions, we can calculate the probability of reaching the requisite two-thirds requirement if ten Republican-nominees and ten Democratic nominees who responded were randomly selected to serve on a real Prosecutor Jury. Using the binomial distribution, we estimate that the probability that at least 14 of a balanced 20-attorney panel would support indictment for Inciting, Assisting or Giving Aid and Comfort to an Insurrection would be only 3.4 percent. In contrast, there is a 76.4 percent chance that at least 14 of a balanced 20-attorney panel would support the Florida indictment.

58

Having said this, we admit that a substantial majority of respondents (86.5%) gave identical responses to each of the four referred charges—either indicating the President Trump should be indicted on all four or indicating that he should not be indicted on any of the four.

Moreover, because of the relative uniformity in responses, the responses did not produce the result where a substantial majority supported some indictment but disagreed about which charge to indict. Accordingly, the “Any Charge” column which reports the proportion of respondents who supported indictment on at least one charge replicates the proportion of the most favored charge of the given nominee group.

59

See, for example, Alschuler (2022), who states that 18 U.S.C. § 2 “declares that someone who aids a crime is a ‘principal.’ She’s guilty of the same crime and subject to the same punishment as the criminal she aids and as someone who commits the crime without assistance.”

60

The differences between the initial and follow up responses for Democratic nominees responding to the Conspiracy to Defraud the United States and Conspiracy to Make a False statement were statistically significant at the 5 percent and 10 percent levels respectively.

61

This result constitutes a violation of the independence of irrelevant alternatives (IIA) criteria. A rational choice axiom prohibits individuals who prefer A to B when given two choices, to switch to preferring B over A when given the choice of A, B, or C (Steele & Stefánsson 2020). The IIA criterion is one of the postulated restrictions on preferences used by Kenneth Arrow in developing his “Impossibility Theorem” (Arrow 1950, p. 337–338).

62

For example, violations of IIA might be caused by “wrong choice” aversion. Imagine that a choice of chocolate or vanilla ice cream is initially offered to a group of people. If the group is then offered the alternative of strawberry ice cream in addition to chocolate and vanilla, it should not cause people to switch from preferring to chocolate to vanilla. But if the group is instead offered the choice between chocolate, vanilla and crappy vanilla, then we might see a larger proportion of the group choosing vanilla, because they are at least sure that vanilla is a better choice than crappy vanilla and will have avoided making the worst possible choice (Ayres 2017; Kelman et al. 1996).

One possibility for the results of Table 4’s second row is that explicitly giving the third option of “Unsure” had the effect of goading Republican nominees toward giving a more decisive answer. When Republican nominees were only given two options, many declined to answer “Yes” or “No.” When given the option of “Unsure,” however, the number declining to answer substantially diminished and instead more were inclined to say the Special Counsel should indict.

63

There is also some evidence that male respondents were more willing to indict than female respondents.

As shown in Table 5, based on the February survey, for each and every charge, male Democratic nominees expressed more indictment support than female Democratic nominees; and male Republican nominees expressed more indictment support than female Republican nominees. For example, with regard to Obstruction of an Official Proceeding, 75 percent of male Democratic nominees supported indictment, while there was only 62.5 percent of support for female Democratic nominees. Based on the July survey, female Democratic nominees expressed more support than male Democratic nominees for the allegations contained in the New York State Indictment; on the other hand, male Republican nominees expressed more support than female Republican nominees. With regards to the Federal Florida Indictment, both Democratic and Republican female nominees expressed more support than their male counterparts. But none of the male/female differences reported in Table 5 are statistically significant at even the 10% level.

Author notes

James Monroe Distinguished Professor of Law, University of Virginia, Gratitude to Darshna Jadhav, Karissa Kang and Nor Ortiz, each of whom assisted us in surveying former US attorneys. Hunter Heck, Karissa Kang, Ben Buell, Aquila Maliyekkal, and Richard Peay supplied excellent research assistance. Thanks to Don Dripps, Ellen Podgor, and Kate Stith, an anonymous reviewer, and seminar participants at ETH Zurich for helpful conversations and comments. Thanks also to Refdesk.

Published by Oxford University Press 2024.