My Profile Search Directory Submit News Contact Us Logout Alumni Home
Fall 2011UVA Lawyer - Home
Dean's MessageOpinionClass NotesIn MemoriamIn PrintFaculty Briefs Home

William J. Stuntz | Criminal "Justice" | Shared Pathologies | Hatching a Gadfly
Prosecution Clinic | Innocence Project | Strip Searches and Immigration Stops

Criminal "Justice" Demanding Certainty in an Uncertain World

prisoner art

by Cullen Couch

“The criminal justice system is characterized by extraordinary discretion—over the definition of crimes (legislatures can criminalize as much as they wish), over enforcement (police and prosecutors can arrest and charge whom they wish), and over funding (legislatures can allocate resources as they wish). In a system so dominated by discretionary decisions, discrimination is easy, and constitutional law has surprisingly little to say about it.”

—William J. Stuntz, The Uneasy Relationship between Criminal Procedure and Criminal Justice

In 1993 at a congressional hearing on capital punishment, an increasingly agitated witness named Federico Macias listened to testimony acknowledging that while innocent persons may be executed, the sentencing statutes provided “unprecedented safeguards” and the risk was “minimal.”

It was not a “mimimal” risk to Macias. Nine years earlier he had been wrongly convicted in Texas of hacking to death Robert and Naomi Haney with a machete during a burglary. Most of the goods stolen from the murdered couple were retrieved from the yard of 19-year-old Pedro Luevanos. Luevanos implicated Macias as his accomplice and as the actual killer. In return Luevanos received a 25-year-sentence while Macias was sentenced to death.

Lucky for Macias, if “lucky” means spending almost a decade on death row, Deirdre Enright ’92, then with Skadden, Arps, took on Macias’s case as member of the firm’s pro bono team of lawyers and investigators. Now director of investigation for the Innocence Project clinic at the Law School, Enright found considerable evidence to support a reversal based on ineffective counsel at trial. The law firm obtained a federal writ of habeas corpus to overturn Macias’s conviction. A unanimous panel of the U. S. Court of Appeals for the Fifth Circuit upheld the writ and found that not only was Macias’s original counsel grossly ineffective, they had also missed considerable evidence pointing to Macias’s innocence.

At the congressional hearing, Macias was among a group of exonerees who told the committee their stories. Macias had twice been so close to execution that he was transferred to the “watch cell,” the cell nearest to the execution chamber. As he listened to the testimony about the “minimal” risk to innocent people, he turned around in his chair and gave his legal team a look of complete disbelief that said, “Are you kidding? An innocent person, like … me?”

Finding Truth in a Labyrinth

Federico Macias’s appalling experience on death row represents the illusory reality of all felony cases in the criminal justice system. Whether a case will proceed is utterly in the control of the prosecutor. Their discretion is close to absolute. If the prosecutor does charge, the case will follow a series of stories told by victims and defendants and witnesses, buttressed by forensic evidence, which build a grand narrative as the case navigates a maze of procedural rules.

The system relies fundamentally on the accuracy and reliability of the data it takes in; “garbage in-garbage out” in computer parlance. From the moment the police handcuff a defendant, information begins to flow into the criminal justice system. Did the police Mirandize the defendant, secure the crime scene, handle the evidence properly, maintain the chain of custody, conduct a proper lineup, obtain a voluntary confession?

At any point during this journey, luck, circumstance, incompetence, or tainted evidence can send the story seriously off the rails. But once “written” and finalized by plea bargain or trial, the story’s “truth” becomes iconic, practically impervious to legal challenge post-conviction.

Like many professions—medicine comes first to mind—the criminal justice system’s skilled practitioners maintain strict standards of conduct, but the stakes are high. Even the smallest error can have ruinous consequences for individuals, families, and communities. The guilty can be freed, or the innocent jailed, and judges have only a modest impact on the process; plea bargains account for 95% of all felony convictions. If the case does go to trial, judges usually rule only on procedural matters.

In fact, most felony cases do produce a legally accurate story. Most defendants do get a fair and impartial hearing. But the criminal justice system, like any human endeavor, suffers from frailty and error. Prosecutorial discretion is its core principle and, at times, its most glaring weakness.

The late William J. Stuntz ’84, former professor at the Law School and seminal thinker on criminal law, often wrote about an unhealthy nexus between prosecutors and legislators that distorts original legislative intent, which Stuntz called the “expressive meaning” of the law.

“What, after all, does expressive criminal law express?” he asked. “Is the message the law that the legislature passes? Or is it the sum of the arrest and prosecution decisions of individual police officers and prosecutors?”

“Prosecutors are better off when criminal law is broad than when it is narrow,” he wrote. “Legislators are better off when prosecutors are better off. The potential for alliance is strong, and obvious. And given legislative supremacy—meaning legislatures control crime definition—and prosecutorial discretion—meaning prosecutors decide whom to charge, and for what—judges cannot separate these natural allies.”

Stuntz’s critique was particularly salient given that he was an evangelical Christian and a political conservative. His views didn’t conform to the usual ideological formulas that define “left” and “right.”

“Bill Stuntz found the problem with the criminal justice system is not so much what we do about the innocent, but what to do with the guilty,” says Law School professor Rachel Harmon, a former federal prosecutor who teaches criminal law, criminal procedure, and civil rights. “The system doesn’t become fair if you mistreat only them. Focusing only on innocence can sometimes distract us from how difficult it is to manage the guilty.”

 By the Numbers

“Criminal punishment is good and proper in the right cases, and incarceration may be a proper form of that punishment …. Still, those of us who live outside those awful warehouses have good reason to be penitent for what we have done to, and failed to do for, those inside.”

—William J. Stuntz, Christian Legal Theory

According to a 2006 Department of Justice report, over 7.2 million people in the United States are either in prison, on probation, or on parole; about one out of every 31 Americans. Approximately 2.5 million are currently incarcerated in state (90%) or federal prison (10%). About 95% of criminal cases are state cases, but longer federal sentences create an outsized number of federal prisoners. The United States leads the world in the number of people it jails. China, with over four times the population, is second, incarcerating 1.6 million.

“We have a political system that puts crime on the front burner of political agendas in a way that’s not true in other countries,” says Darryl Brown ’90, a former public defender and now the Law School’s O.M Vickers Professor of Law. “That makes it more likely that we criminalize too much. Our punishment levels are five to seven times higher than those of any industrialized European countries.”

Do those punishment levels have any effect on the crime rate? “Americans commit more violent crimes than citizens of any European country, and about the same amount of property crime, non-violent crime, and drug crime. Clearly we aren’t getting more bang for the buck,” says Brown.

Brown would like to see a return at least to the sentencing levels the U.S. had in the early 1980s, before the “war on drugs” prompted increasingly severe sentencing policies. “Our incarceration policies were harsher in the early ‘80s than they had been for most of the twentieth century, and more severe than Europe has ever had,” he says. “So that was hardly a lenient policy. But I just don’t think there’s any evidence that shows we’re improving safety or reducing crime from the ever-longer sentences we’ve added in the last fifteen to twenty years. And incarceration is incredibly costly financially, holding aside the human costs.”

“The relationship between punishment and crime is hard to establish,” adds Harmon. “As a society, we’re doing something deeply wrong.”

Further, since between one and five percent of guilty verdicts are incorrect, although it is virtually impossible to know for sure, too many innocent are being punished, but a far greater number suffer from the system’s mistreatment of the guilty.

“If you spend time in prisons,” says Harmon, “you find that you can’t imagine the problem to which prisons are the solution, unless you’re simply talking about not allowing people to be free to commit more crimes. But as a way of remedying the social problem of crime, it’s abhorrent. Our prison system is deeply dysfunctional.”

On the Street

“We are likely to come ever closer to a world in which the law on the books makes everyone a felon, and in which prosecutors and the police both define the law on the street and decide who has violated it.”

—William J. Stuntz, The Pathological Politics of Criminal Law

“We want policing to work,” says Harmon. “We want it to reduce fear, to reduce disorder, and to reduce crime. It’s very easy to be critical about police use of excessive force and forget that we don’t presently have an alternative to policing to achieve these ends.”

There are approximately 17,000 police departments in the U.S., from the enormous municipal departments of major cities like New York (35,000) to the smallest two-person sheriff’s office in thousands of rural communities. Overall, there are close to 900,000 law enforcement officers in the U.S.

The vast majority of police officers work in intensely supervised environments, according to Harmon, who has prosecuted police for using excessive force.

The question is how to regulate policing to achieve social order at the lowest cost. For example, SWAT teams are much more intrusive than Terry stops (brief detention on reasonable suspicion), but each of them can cause harm. Are they worth it?

“In New York, police make more than 500,000 Terry stops a year,” Harmon says. “Those numbers are astronomical. Critics say police rarely find anything and they’re often violating constitutional rights. The police say they rarely violate constitutional rights and they often find something. That debate ignores that we have to weigh the benefits—even when Terry stops are constitutional and effective—against the aggregate dignity harm of stopping that many people on the street.”

Nor can one know if the presence of a SWAT team during a search increases the likelihood of civilian or police fatalities, she says. These are often high risk situations where the occupants are very likely to be armed. Does a SWAT team protect or injure more people?

In both cases, there is simply no empirical evidence to know.

To further complicate the issue, institutional problems inside police departments can be hidden by an administrative system that can easily scapegoat a few bad actors, wash its hands, and resume normal operations.

 “One of the most disturbing things I realized when I was prosecuting police officers was that instead of fixing the underlying problem of police misconduct, I was sometimes making it worse,” Harmon says. “Some of the people I prosecuted were bad people who did terrible things, but I also knew from my investigations that departments sometimes tolerated the bad behavior. Yet when federal prosecutors finish a case, the department chief often holds a press conference, thanks the Justice Department for taking care of the problem, and says everything can go back to normal.”

Rick Moore ’80, Assistant Commonwealth Attorney for Orange County, Virginia, and director of the Law School’s Prosecution Clinic, understands these concerns, but cautions that the necessary focus on the relatively small number of “bad actors” skews the reality. “If something really bad occurs in one percent or even five percent of cases, and that’s all a law student studies, they develop a jaded view of the police and prosecution because they really haven’t seen what’s out there. What you don’t hear about are the hundreds and thousands of things that happen every day where somebody’s been courteous or fair or transparent. That doesn’t get any press.”

Doug Gansler ’89, Attorney General of Maryland and a former county and federal prosecutor, agrees. “Most police officers are in it for the right reasons. They want to protect the community and they’re woefully underpaid to do that. Of course, some are better police officers than others. They’re going to go the extra mile to get better evidence to make a better case. They will look for their extra eyewitness or will to do a better job at the crime scene.”

At the Station House

“Criminal procedure’s rules and remedies are embedded in a larger system, a system that can adjust to those rules in ways other than obeying them. And the rules can in turn respond to the system in a variety of ways, not all of them pleasant. The more one focuses on that dynamic, the more problematic the law of criminal procedure seems.… Ever since the 1960s, the right has argued that criminal procedure frees too many of the guilty. The better criticism may be that it helps to imprison too many of the innocent. ”

—William J. Stuntz, The Uneasy Relationship between Criminal Procedure and Criminal Justice

According to a congressionally mandated report from the National Academy of Sciences released in February 2009, even if the police perform their role to the highest professional standards, the validity of the data that flows within this closed system can still have “serious deficiencies.”

“With the exception of nuclear DNA analysis…, no forensic method has been rigorously shown able to consistently, and with a high degree of certainty, demonstrate a connection between evidence and a specific individual or source,” says NAS. The report concluded that the forensic system is in a “nationwide crisis” and needs to be completely overhauled.

“Forensic evidence appears really powerful to people,” says Brandon Garrett, the Law School’s Roy L. and Rosamund Woodruff Morgan Professor of Law, who presented to the NAS committee. “It turns out, though, that much of that evidence is of unknown and untested reliability. Basic research still needs to be done for us to know what those conclusions really mean and whether the evidence really tells us what we’d like to know.”

Some of the basic techniques identified in the NAS report included fingerprint comparison, fiber comparison, hair comparison, and ballistics comparison, all quite commonly used. Each contains detailed information, but that does not mean that there is any scientific support for analysts’ claims made confidently in reports and at trial that they can tell that the evidence came from the defendant.

Following up on his NAS analyses, Garrett conducted in-depth research of the trial transcripts of 250 exonerated defendants, and published his findings in his recent book, Convicting the Innocent: Where Criminal Prosecutions Go Wrong. The study provided a unique chance to start with a certainty and work backwards to see how that error occurred.

“We knew that these trials convicted innocent people, “he says, “so I wanted to read in detail the records from those cases and see what could be learned from that. What came out of it at the end of the day were some pretty distinct patterns and some real opportunities for changing criminal procedural rules.”

In each of these cases, the forensics seemed powerful, the evidence strong, the confessions real, and the eyewitnesses confident. But they all led to the wrong verdict. “The jury didn’t see the evidence as it really was,” Garrett found. “The evidence was not carefully documented and it was often contaminated.”

Examiners have long asserted that every fingerprint, every bullet casing, and every bullet is unique without testing those propositions, according to Garrett. “They have made exaggerated claims without scientific support. We just have no idea the frequency of particular markings on a fingerprint, and so we have no idea what it means for an examiner to say that fingerprints match.”

“As the NAS describes—and it is a powerful report well worth reading—there are a host of problems with these forensics,” concludes Garrett. “If research was done correctly and findings presented in a more cautious way, they could provide a sound source of evidence, but judges haven’t demanded much of forensic analysts. They’re given a free ride to present unreliable forensics in a misleading way.”

 Eyewitness testimony is crucial in tens of thousands of cases each year, and while we all are familiar with how hard it is to recognize the faces of strangers, many do not realize how powerfully police lineup procedures can affect the confidence of a witness or even lead them to identify the wrong person. Fully three-quarters of the exoneree trials Garrett studied involved eyewitness testimony, and many cases had not just one, but two, three, or more eyewitness. Almost all of the eyewitnesses told the jury that they were absolutely certain of their identification. All of them were wrong.

 False confessions are particularly disturbing. “I think we all have a visceral reaction to false confessions,” says Garrett. “It is an outright nightmare for an innocent person to be placed in an interrogation room, where the goal of the police is to get a confession.”

Still, why would someone confess to something that they didn’t do? Some innocent people may make a rational choice that the only way to escape the interrogation room is to play along with the police. “They may want to get out of this horrendous, grueling situation,” says Garrett. “They may have figured that since they’re innocent, they can clear it up later. They think, ‘I’ll just repeat after them and say what they tell me to say; it’s all a joke anyway. I’ll prove my innocence to the judge.’” Psychologists call that phenomenon a coerced compliance confession.

In other cases, which psychologists call coerced internalized confessions, the stress of an interrogation can convince an individual that they are guilty when they are not. Police may tell them that maybe they blacked out, or didn’t remember. “A particularly vulnerable person,” says Garrett, “can actually become convinced that they did something that they never did.”

Courts also allow police to lie during interrogations. It’s an approved practice. “Police can lie about almost anything,” says Enright. “Lie about what they know, lie about what evidence they have, lie about what witnesses have—or have not—told them. How can anyone think that feeding people false or misleading information will somehow generate the truth?”

Matthew Engle, director of the Innocence Project Clinic at the Law School, says interrogators often tell suspects they are free to go, if.… “They say, ‘Tell me now that you did it, and then you can go home. You can walk out the door. This can all end right now.’ As soon as that so-called confession is signed, it’s game over. If the defendant doesn’t invoke his right to counsel immediately, this happens all the time.”

In the cases handled by Engle and Enright, police have given Miranda warnings orally and then asked the defendant to initial a piece of paper acknowledging that they have the right to remain silent. But then, “nobody invokes their right to counsel,” says Engle. Engle and Enright can recall only three or four clients during their careers who remained silent after being read their rights.

“Our clients tell us they didn’t have anything to hide,” says Enright. “They talked because they didn’t have any reason not to. Then the interrogator begins lying. The interrogator turns the tape recorder on and off and then perhaps they all go outside for the ‘I don’t know anything’ cigarette break. And then, when they come back the defendant suddenly knows everything there is to know about the crime.”

Admittedly, these are extreme examples, and perhaps support Moore’s point that they unfairly color a criminal justice system that overwhelmingly gets it right. Certainly, the Innocence Project handles those cases in which an extreme injustice is apparent. That’s the clinic’s norm. But the record is not so clear in Garrett’s study of the exoneree cases.

“If anything, what is most disturbing is that in most jurisdictions, we are still using many of the same procedures used in those innocent people’s cases,” he says. “It’s not that hard or expensive to collect better evidence, to adopt scientific standards for presenting the forensics, to videotape the confessions, to make sure that the lineups are double-blind. These are all easy steps to take. Slowly, but hopefully surely, we will start to see real criminal procedure change.”

Garrett credits prosecutors in many jurisdictions around the country who have become an important force in improving criminal investigation. Prosecutors know that jurors are more aware of the problems with eyewitness memory. They want to tell the jury that they used double-blind lineups that prevented the possibility of suggestion. They want to videotape full interrogation sessions to show the jury nothing unprofessional happened.

“Wrongful convictions are a civil rights problem and a law enforcement problem. After all, in each of the cases I studied, the guilty remained free. Dozens committed additional rapes and murders before DNA testing freed the innocent and brought them to justice,” says Garrett. “Improving the accuracy of our criminal justice system is important to make sure we get it right the first time.”


“Legislators gain when they write criminal statutes in ways that benefit prosecutors. Prosecutors gain from statutes that enable them more easily to induce guilty pleas. Appellate courts lack the doctrinal tools to combat those tendencies.”

—William J. Stuntz, The Pathological Politics of Criminal Law

Effective regulation of prosecutorial discretion rests with administrative agencies in each state, the federal government, and internal controls in individual prosecutors’ offices. The U.S. has essentially 51 jurisdictions, all but guaranteeing inconsistencies throughout, not only in regulating prosecutors but also in charging and sentencing.

Commit a crime in Maryland, for example, and your sentence would be lighter than if you had committed the same crime across the river in Virginia. “Is one right or one wrong?” asks Gansler. “Probably not. It just reflects the mores of the two different jurisdictions.”

Those differences came into stark relief in the case of Troy Davis in Georgia. Davis was executed for a murder he claimed he did not commit. His conviction was based solely on eyewitness testimony, notorious for its inaccuracy. The Georgia State Board of Pardons and Paroles and the Governor refused Davis’s clemency application. Had Davis been convicted in Maryland, he could not have received the death penalty; a Maryland statute enacted in 2009 prevents the use of the death penalty on convictions obtained solely through eyewitness testimony.

Many other countries have centralized police and prosecutors. There are detailed rules they all must obey. “But that’s just not who we are,” says Harmon, “and there are real consequences to that. Do prosecutors have a lot of power? Sure they do.”

Gansler turns the issue of discretion around, particularly for new lawyers interested in protecting the innocent from false accusations. “You learn pretty quickly in our criminal justice system that the prosecutor has all the power,” he says. “If you’re inclined toward fairness in the system, then you learn that prosecution is where you should be. For example, if you’re confronted with an innocent defendant, you can make sure that person is released.”

Moore says the ideals of public service drive most prosecutors. “You’re helping uphold or enforce the community’s rules of safety and welfare. When people break the community’s rules, someone’s got to hold them accountable, and do it fairly. It’s the type of job where you can do what you think is right almost every day.”

State prosecutors handle approximately 95 percent of so-called ordinary crime: thefts, low to middle-level drug offenses, and violent crime. With limited resources, they tend to prosecute the most dangerous defendants who cause the most severe harm. Within those constraints, they have almost complete autonomy over what they choose to pursue.

“I don’t know of a case where you don’t have discretion,” says Moore. “We use our discretion not to prosecute—even a vicious murder—if we don’t think the evidence the police obtained is strong enough.”

State prosecutors can also target specific types of criminal activity that is damaging the community. When Gansler was the district attorney for Maryland’s Montgomery County, he started a special unit to combat criminal gang activity. Gansler appealed to his representative to bring a bill to the Maryland General Assembly to improve his ability to prosecute gang cases. When Gansler became the state’s Attorney General, he created a state-wide gang unit.

“That was a policy decision,” says Gansler. “Gangs are hurting kids and you can’t learn if you’re in school with gang members sitting next to you. We needed to address the gang problem in a different way than had previously been done.”

Federal prosecutors focus primarily on high-profile white collar crime, public corruption, organized crime, and inter-state and international drug trafficking operations. With so many federal statutes on the books, federal prosecutors exercise more discretion than state prosecutors. Working with a staff of 127 assistant U.S. attorneys, Zane Memeger ’91, U.S. Attorney for the Eastern District of Pennsylvania, has organized his office into two main sections, criminal and civil. The criminal division includes units on economic crime, public corruption and civil rights, government and health care fraud, violent crime, narcotics, cyber-crime, organized crime, and terrorism.

“It’s a question of resources,” says Memeger. “There are a lot of statutes on the books that we can enforce, and Philadelphia and its surrounding eight counties have a wide variety of issues that we need to address. Terrorism is the no. 1 priority for the Department of Justice, but putting that aside, we have industries in the area that create an environment for economic crime and corruption—financial, health care, education, insurance, shipping, and others. We also have 5.5 million people in our district, and we have significant violence issues involving drugs and armed robberies. Violent crime is generally on a downward trend throughout most of the nation, but not here in Philadelphia. Compared to other major cities, the shooting and murder rates remain consistently high.”

Like the assistant U.S. Attorneys in Memeger’s office, Harmon had a great deal of oversight. She couldn’t initiate a grand jury proceeding without the approval of a supervisor and assistant attorney general. She couldn’t bring an indictment or take a plea without approval.

“The picture of prosecutors as unconstrained actors in a system that simply empowers them needs to be refined for different areas of law,” she says. “It often doesn’t make sense to think about bad prosecutors or good prosecutors. I never met anyone trying to convict innocent people of crimes or to deny defendants their rights.”


“Perhaps sacrifice is an essential part of doing justice for the poor. Perhaps the key is not simply to generate accurate outcomes, but somehow to enter into the defendant’s distress. Burlette Carter’s idea, that Christianity requires seeing the world through the other’s eyes and not one’s own, may apply best to the world of criminal defense, where empathy is most needed yet seems to be in shortest supply.”

—William J. Stuntz, Christian Legal Theory

Anyone who spends a day in criminal court has to wonder how the system can work at all. The prosecutor stands by a stack of files, pulling one case after another off the top, calling a name, and then looking at it for the first time. Usually, neither the prosecutor nor the defense attorney nor the judge knows anything about the case. Nevertheless, they start negotiating plea bargains, guilty pleas, and even conduct short trials using only the information at hand. It would be extraordinary if they handled each case correctly.

“The defense attorney often has never met his or her client,” says Engle. “Not because they are lazy or incompetent, though that can happen, but because they’re overworked in a system with an enormous volume of cases.”

“One of the best solutions to the problems of the criminal justice system,” says Harmon, “is to fund defense lawyers better. One of the reasons that the innocence mistakes aren’t picked up sooner is because they didn’t have a Deirdre Enright the first time around. Criminal defendants just don’t have the resources they need.”

Flawed forensic evidence often goes unquestioned, adds Garrett. “Defense lawyers simply can’t afford their own experts and judges may not let them in anyway. If they don’t ask very many questions about the science, it’s because they don’t understand it without their own expert advice.”

A common complaint by the defense bar, especially in Virginia, is that prosecutors are not required to have an “open file” policy regarding discovery. By Virginia statute, a prosecutor must give the defense only exculpatory evidence, their client’s statement, and forensics reports. Although most Commonwealth’s attorneys go beyond that minimal requirement and have an “open file” policy, it’s not required.

Clearly, it is difficult to mount an effective defense if prosecutors never disclose the interactions between police and eyewitnesses, crime lab analysts, informants, or others which occurred before the lawyer even became involved. “If evidence is routinely undocumented, confessions are not videotaped, lab notes are not disclosed, lineups are not recorded, how can lawyers and judges and jurors sort out what went wrong?” says Garrett. “I found a disturbing number of cases where police and prosecutors concealed evidence of innocence—in cases where DNA later proved the defendant’s innocence. We will never know if still more evidence remains concealed to this day. And if there was so much misconduct in those very serious criminal trials, imagine what evidence never comes to light in the vast majority of cases where there is never a trial, but rather a plea bargain?”

The closed file policy turned out to be a very big problem in one of the Innocence Project’s cases involving the murder conviction of Justin Wolfe (see page 44). It took a decade—and a federal district court order—to compel the prosecution to open its files. When they were finally produced, Enright and Engle found a trove of evidence that might have exonerated Wolfe had it been discovered before trial. The judge agreed and vacated the verdict. Engle asks, “What does that suggest about the other 95% of the cases where that kind of scrutiny never takes place because discovery never happens?”

Moreover, why would the prosecution not turn over its files on request, especially in a death penalty case that draws such scrutiny? Some prosecutors might have convinced themselves that the defendant is guilty and if they give the defense this random piece of information, the defendant will use it to manufacture an illegitimate defense. “It betrays ignorance about how the system should work,” says Harmon, “but also a deep belief by the prosecutor that he is justified in doing it. That’s almost more disturbing than if evil prosecutors are actually out there doing the wrong thing on purpose.”

Plea, or Take a Chance at Trial?

“Broad criminal codes ensure inconsistency. Broad codes cannot be enforced as written; thus, the definition of the law-on-the-street necessarily differs, and may differ a lot, from the law-on-the-books.”

—William J. Stuntz, The Pathological Politics of Criminal Law

The sheer volume of cases forces the criminal justice system to rely largely on the plea bargain which, like the rest of the system, is entirely dependent on the quality and fairness of prosecutors. They set the terms of the debate, especially when they have a whole host of offenses to use. If they want, they can “stack” together multiple, overlapping offenses against a defendant—effectively picking sentences as well, since they are locked in to some charges—making trial a risky proposition even if the defense thinks they have a good chance of winning.

“Dramatic differences between trial and plea sentences put a lot of pressure on defendants facing those kinds of charges,” says Brown. “They often feel they need to plead guilty to things that they didn’t do. It can really skew their judgment.”

Unlike much of the rest of the world, the American criminal justice system does not regulate the discount the defendant can get between a plea bargain sentence and a sentence after trial. Every plea bargaining system that Brown has studied has a rule that prohibits discounts any larger than one-third of the sentence that a defendant would get after trial.

That would effectively limit the leverage a prosecutor can use against defense during plea negotiations. Instead, in a case that Stuntz wrote about in “Bordenkircher v. Hayes: The Rise of Plea Bargaining and the Decline of the Rule of Law,” the Supreme Court ruled that the prosecution can pose as dramatic a contrast between the plea outcome and the post-trial sentence as the statutes will allow. In that case, the prosecution offered two or three years for the plea bargain and, when the defendant turned it down, took the case to trial. The defendant was convicted and received a mandatory sentence of life in prison.

Further, defendants convicted by plea bargaining may not receive appropriate charges or sentences. “The prosecutor’s going to have a different idea than the defense of what’s fair,” says Brown. “The public also may have a different idea of what’s an appropriate resolution, but there is no real public supervision by the judge of the prosecution’s discretion and judgment.”


“Appellate criminal litigation used to be primarily substantive; the focus was on either the sufficiency of the evidence or the definition of crimes or defenses. Today it is overwhelmingly procedural; Fourth, Fifth, and Sixth Amendment claims have taken the place of substantive claims.”

—William J. Stuntz, The Pathological Politics of Criminal Law

Post-conviction challenges face daunting obstacles, for good reason. Judges are reluctant to draw conclusions from a cold record. They look rightly to the jury as the finder of facts. They are also time-consuming and expensive to reinvestigate and relitigate. But when there is a compelling reason to do so, and an ability to dedicate resources to the effort, good lawyers and investigators can introduce evidence that never arose at trial.

Prosecutors naturally react defensively when challenged post-conviction by new evidence or a recanting witness. “They have confidence in their case,” says Harmon. “Sometimes it’s personal bias, but they also know the costs to the victims, the victim’s family, and to the community that arise from reopening criminal matters.”

Moore says prosecutors also want to ensure that they don’t wrongly vacate a proper conviction. “Prosecutors have a vested interest in the original outcome, but if they’re confronted with evidence they think really calls into question the original result, they’ll pursue that. But if they think this new information does not challenge their belief in the evidence of guilt, they’re going to fight you all the way.”

And that is a fight very difficult for a defense attorney to win. “The rules in post-conviction are very restrictive,” says Engle. “There are many procedural hurdles and roadblocks that prevent inmates from getting access to the information they need to attack a conviction, much less to have courts hear it and consider it and to prevail.”

Antonin Scalia’s dissent in Herrera shows the extreme height of those hurdles. “There is no basis in text, tradition, or even in contemporary practice (if that were enough), for finding in the Constitution a right to demand judicial consideration of newly discovered evidence of innocence brought forward after conviction.” To Scalia, procedure takes precedence over substance, even in a capital case.

It is an attitude that pervades the system. In his study of the exoneree trials, Garrett revealed a tortured path to freedom. It was exceedingly difficult for the convicted to get access to the DNA that ultimately established their innocence. Even with that, they had to wait years before being released. “It certainly doesn’t speak well that our system couldn’t respond quickly even to the easy cases where DNA shed clear light on their innocence,” he says.

Too often, says Enright, “We’re forced into these absurd positions. Why can’t a prosecutor just admit it was a mistake? They thought it was the defendant. Some tests showed they were wrong. Our criminal justice system, and all the players in it, has to be able to accommodate error and mistake, because we all know that we make them.”

The Innocence Project can spend years on a case. Dozens of students do the legwork. The clinic has all the advantages that none of the other players in the system had when it mattered. “Most criminal cases don’t get that kind of scrutiny post-conviction,” says Garrett. “You wonder how many others have the same problems.”

“We give them the presumption of innocence after they’ve been found guilty, because usually, nobody else did,” says Enright. “We start with the presumption that the potential client is telling the truth, and it’s disturbing how often they are.”

Certainty in law is the cornerstone of a stable society. But certainty itself can be a brittle thing. “Theories of everything, if they are at all plausible, are deep and rich and complex, so much so that none of us can finally get to the bottom of them,” warned Stuntz. “So the things the theory helps explain can also help explain the theory. Truth runs in both directions.”