Prosecutorial discretion poses an increasing threat to justice. The threat has in fact grown more severe to the point of becoming a due process issue. Two recent events have brought more attention to this problem. One involves the decision not to charge NBC anchor David Gregory with violating gun laws. In Washington D.C., brandishing a thirty-round magazine is illegal and can result in a yearlong sentence. Nonetheless, the prosecutor refused to charge Gregory despite stating that the on-air violation was clear.1 The other event involves the government’s rather enthusiastic efforts to prosecute Reddit founder Aaron Swartz for downloading academic journal articles from a closed database. Authorities prosecuted Swartz so vigorously that he committed suicide in the face of a potential fifty-year sentence.2
Both cases have aroused criticism. In Swartz’s case, a congresswoman has even proposed legislation designed to ensure that violating a website’s terms cannot be prosecuted as a crime.3 But the problem is much broader. Given the vast web of legislation and regulation that exists today, virtually any American bears the risk of being targeted for prosecution.
I. The Problem with Prosecutorial Discretion
Attorney General (and later Supreme Court Justice) Robert Jackson once commented: “If the prosecutor is obliged to choose his cases, it follows he can choose his defendants.”4 This method results in “[t]he most dangerous power of the prosecutor: that he will pick people he thinks he should get, rather than pick cases that need to be prosecuted.”5 Prosecutors could easily fall prey to the temptation of “picking the man, and then searching the law books . . . to pin some offense on him.”6 In short, prosecutors’ discretion to charge—or not to charge—individuals with crimes is a tremendous power, amplified by the large number of laws on the books.
Prosecutors themselves understand just how much discretion they enjoy. As Tim Wu recounted in 2007, a popular game in the U.S. Attorney’s Office for the Southern District of New York was to name a famous person—Mother Teresa, or John Lennon—and decide how he or she could be prosecuted:
It would then be up to the junior prosecutors to figure out a plausible crime for which to indict him or her. The crimes were not usually rape, murder, or other crimes you’d see on Law & Order but rather the incredibly broad yet obscure crimes that populate the U.S. Code like a kind of jurisprudential minefield: Crimes like “false statements” (a felony, up to five years), “obstructing the mails” (five years), or “false pretenses on the high seas” (also five years). The trick and the skill lay in finding the more obscure offenses that fit the character of the celebrity and carried the toughest sentences. The, result, however, was inevitable: “prison time.”7
With so many more federal laws and regulations than were present in Jackson’s day,8 a prosecutor’s task of first choosing a possible target and then pinning the crime on him or her has become much easier. If prosecutors were not motivated by politics, revenge, or other improper motives, the risk of improper prosecution would not be particularly severe. However, such motivations do, in fact, encourage prosecutors to pursue certain individuals, like the gadfly Aaron Swartz, while letting others off the hook—as in the case of Gregory, a popular newscaster generally supportive of the current administration.
This problem has been discussed at length in Gene Healy’s Go Directly to Jail: The Criminalization of Almost Everything9 and Harvey Silverglate’sThree Felonies a Day.10 The upshot of both books is that the proliferation of federal criminal statutes and regulations has reached the point where virtually every citizen, knowingly or not (usually not) is potentially at risk for prosecution. That assertion is undoubtedly true, and the consequences are drastic and troubling.
The result of overcriminalization is that prosecutors no longer need to wait for obvious signs of a crime. Instead of finding Professor Plum dead in the conservatory and launching an investigation, authorities can instead start an investigation of Colonel Mustard as soon as someone has suggested he is a shady character. And since, as the game Wu describes illustrates, everyone is a criminal if prosecutors look hard enough, they are guaranteed to find something eventually.
Overcriminalization has thus left us in a peculiar place: Though people suspected of a crime have extensive due process rights in dealing with the police, and people charged with a crime have even more extensive due process rights in court, the actual decision of whether or not to charge a person with a crime is almost completely unconstrained. Yet, because of overcharging and plea bargains, the decision to prosecute is probably the single most important event in the chain of criminal procedure.
II. Checks on Prosecutorial Discretion
Despite the problems described above, most of us remain safe. Prosecutors have limited resource and there are political constraints on egregious overreaching.11 And presumably, most of the time prosecutors can be expected to exercise their discretion soundly. Unfortunately, these limitations on prosecutorial power are likely to be least effective where prosecutors act inappropriately because of politics or prejudice. Limited resources or not, a prosecutor who is anxious to go after a political enemy will always find sufficient staff to bring charges, and political constraints are least effective where a prosecutor is playing to public passions or hysteria.12
Once charged with a crime, defendants are in a tough position. First, they must bear the costs of a defense, assuming they are not indigent. Second, even if they consider themselves entirely innocent, they will face strong pressure to accept a plea bargain—pressure made worse by the modern tendency of prosecutors to overcharge with extensive “kitchen-sink” indictments: Prosecutors count on the fact that when a defendant faces a hundred felony charges, the prospect that a jury might go along with even one of them will be enough to make a plea deal look attractive. Then, of course, there are the reputational damages involved, which may be of greatest importance precisely in cases where political motivations might be in play. Worse, prosecutors have no countervailing incentives not to overcharge. A defendant who makes the wrong choice will wind up in jail; a prosecutor who charges improperly will suffer little, if any, adverse consequence beyond a poor win/loss record. Prosecutors are even absolutely immune from lawsuits over misconduct in their prosecutorial capacity.13
III. Better Approaches to Prosecutorial Accountability
So how to respond? Although this brief Essay cannot begin to address all of the possibilities, it can serve as the beginning of a much-needed discussion. As this Essay indicates, the decision to charge a person criminally should itself undergo some degree of due process scrutiny. Short of constitutional due process scrutiny, however, it is time to look at structural changes in the criminal justice system that will more successfully deter prosecutorial abuse.
Traditionally, of course, the grand jury was seen as the major bar to prosecutorial overreaching.14 The effectiveness of this approach may be seen in the longstanding aphorism that a good prosecutor can persuade a grand jury to indict a ham sandwich.15 Grand jury reforms—where grand juries still exist—might encourage grand jurors to exercise more skepticism and educate them more.16 But grand juries are not constitutionally guaranteed at the state level, and reforming them at the federal level is likely to prove difficult.
Overall, the problem stems from a dynamic in which those charged with crimes have a lot at risk, while those doing the charging have very little “skin in the game.” One source of imbalance is prosecutorial immunity. The absolute immunity of prosecutors—like the absolute immunity of judges—is a judicial invention, a species of judicial activism that gets less attention than many other less egregious examples. Although such immunity no doubt prevents significant mischief, it also enables significant mischief by eliminating one major avenue of accountability. Even a shift to qualified, good faith immunity for prosecutors would change the calculus significantly, making subsequent review something that is at least possible.
Another remedy might be a “loser pays” rule for criminal defense costs. After all, when a person is charged with a crime, the defense—for which non-indigent defendants bear the cost—is an integral part of the criminal justice process.17 For guilty defendants, one might view this cost as part of the punishment. But for those found not guilty, it looks more like a taking: Spend this money in the public interest to support a public endeavor, or go to jail. Perhaps the prosecution could be required to pay a defendant’s legal fees if he or she is not convicted. To further discipline the process, one could implement a pro-rate system: Charge a defendant with twenty offenses, but convict on only one, and the prosecution must bear 95% of the defendant’s legal fees. This would certainly discourage overcharging.
The “nuclear option” of prosecutorial accountability would involve banning plea bargains. An understanding that every criminal charge filed would have to be either backed up in open court or ignominiously dropped would significantly reduce the incentive to overcharge. It would also drastically reduce the number of criminal convictions achieved by our justice system. But given that America is a world leader in incarceration, it is fair to suggest that this might be not a bug, but a feature.18 Our criminal justice system, as presently practiced, is basically a plea bargain system with actual trials of guilt or innocence a bit of showy froth floating on top.19
A less dramatic option might be to require that the prosecution’s plea offers be presented to a jury or judge after a conviction, before sentencing. Judges or jurors might then wonder why they are being asked to sentence a defendant to twenty years without parole when the prosecution was willing to settle for five. Fifteen years in jail seems a rather stiff punishment for making the state undergo the bother of a trial.
It is also worth considering whether mere regulatory violations—malum prohibitum rather than malum in se—should bear criminal sanctions at all. Traditionally, of course, citizens have been expected to know the law. Yet traditionally, regulatory crimes usually applied only to citizens in specialty occupations, who might be expected to be familiar with applicable regulatory law. Ordinary citizens needed no special knowledge to avoid committing rape, robbery, theft, etc. But now, with the explosion of regulatory law, every citizen is at risk of criminal prosecution for crimes that, as David Gregory’s defenders noted,20 involve no actual harm or ill intent. Yet any reasonable observer would have to conclude that actual knowledge of all applicable criminal laws and regulations is impossible, especially when those regulations frequently depart from any intuitive sense of what “ought” to be legal or illegal. Perhaps placing citizens at risk in this regard constitutes a due process violation; expecting people to do (or know) the impossible certainly sounds like one.
Support for this notion comes from Court of Appeals Judge John R. Brown, who wrote, in a 1965 case holding that a prosecutor could refuse to sign a grand jury’s indictment, that such a refusal was justified by the complexities of modern criminal law:
Putting aside these factors which bear on the delicate nature of governmental decisions, there are technical reasons indigenous to criminal law which are equally compelling. Federal crimes are more and more for violation of highly complex statutes. Federal jurisdiction, indeed, whether the activity constitutes a federal crime, depend on intricate facts, many beyond the knowledge and experience of laymen composing the Grand Jury.21
This naked admission that federal criminal law is so complex that a grand jury cannot be expected to understand it carries two lessons: First, it seems optimistic to expect grand juries to provide an adequate check on prosecutorial overreaching; and second, if a federal grand jury cannot be expected to understand the complexities of federal criminal law, it seems utterly absurd to maintain the fiction that ordinary citizens should be presumed to know the law.
That being the case, it seems to me that the problem here is a real one. If we care about due process—and we should—we should be deeply concerned about a system in which official discretion reigns almost unfettered where constraint matters most.
* Beauchamp Brogan Distinguished Professor of Law, University of Tennessee. J.D., Yale Law School, 1985. Thanks to Leigh Outten for excellent research assistance.
There are now more than 4,000 federal crimes, an increase of one-third since 1980. Many of those crimes, spread out through some 27,000 pages of the U.S. Code, incorporate violations of federal regulations that are in turn spread throughout the tens of thousands of pages of the Code of Federal Regulations. As a result, even teams of legal researchers—let alone ordinary citizens—cannot reliably ascertain what federal law prohibits.
Gene Healy, Go Directly to Jail: The Criminalization of Almost Everything, at vii (2004).
11State prosecutors are often elected and thus subject to direct political constraint. But even federal prosecutors are subject to supervision by the Attorney General or the President, who must take account of public reaction. After Aaron Swartz’s suicide, for example, Justice Department officials were called to explain his prosecution before Congress. Ryan J. Reilly, DOJ to Brief Congress on Aaron Swartz Prosecution, Huffington Post: Politics (Feb. 15, 2013, 12:12 pm), http://www.huffingtonpost.com/2013/02/15/aaron-swartz-prosecution_n_2695356.html (on file with the Columbia Law Review).
13See generally David Keenan et al., The Myth of Prosecutorial Accountability After Connick v. Thompson: Why Existing Professional Responsibility Measures Cannot Protect Against Prosecutorial Misconduct, 121 Yale L.J. Online 203, 209–20 (2011),http://yalelawjournal.org/2011/10/25/keenan.html (on file with the Columbia Law Review) (discussing prosecutorial immunity).
14See, e.g., United States v. Cox, 342 F.2d 167, 170 (5th Cir. 1965) (“The constitutional requirement . . . of an indictment . . . has for its primary purpose the protection of the individual from jeopardy except on a finding of probable cause by a group of his fellow citizens, and is designed to afford a safeguard against oppressive actions of the prosecutor or a court.”).
21United States v. Cox, 342 F.2d 167, 182–85 (5th Cir. 1965) (Brown, J., concurring).
Preferred Citation: Glenn Harlan Reynolds, Ham Sandwich Nation: Due Process When Everything Is a Crime, 113 Colum. L. Rev. Sidebar 102 (2013), http://www.columbialawreview.org/ham-sandwich-nation_Reynolds.