Skip navigation

America's Guilt Mill

by David J. Krajicek, The Crime Report

Thousands of Americans, many of them poor, are wrongfully convicted each year for crimes that don’t make headlines. While innocence advocates focus on lifers, those falsely accused of lesser crimes are the overlooked casualties of our overburdened courts.

When Rachel Jernigan was falsely accused of robbing a Gilbert, Ariz., bank 15 years ago, she expected the American criminal justice system to do the right thing.

“They tried to get me to plead guilty,” Jernigan says. “They told me they were going to give me 27 years (in prison). But I said I’m not going to plead guilty for something I didn’t do. I really believed I was going to come home from my trial. I was shocked when the jury found me guilty.”

Sentenced to 14 years, she spent more than seven years in prison before the real robber was identified by Jernigan’s determination and a fluke twist.

“If it can happen to me,” Jernigan says, “it can happen to anyone.”

And it does. In a sense, Jernigan was a lucky exception. Experts believe that thousands of people are wrongfully convicted each year in America for the types of crimes that Jernigan was charged with—second-tier felonies like robbery, burglary and assault. And when misdemeanors and driving infractions are included, the number of flawed convictions increases exponentially. Yet only a tiny fraction of these cases are ever exposed. The cadre of criminologists and law professors who study wrongful convictions regard these missing exonerations as one of the great mysteries of American criminal justice.

Many believe the victims are likely the low-hanging fruit of the justice machine, poor men and women who don’t have the wherewithal to pursue justice. They likely do what Jernigan was not willing to do: suck it up and accept a plea deal.

“My own somewhat unstudied, seat-of-the-pants estimation is that a lot of working-class folks are probably pretty cynical about the world,” says Marvin Zalman of Wayne State University, a leading wrongful convictions scholar. “And I think that when they get convicted of relatively minor stuff where they didn’t do anything wrong, they just chalk it up to a bad experience, do their time, and simply move on.”

Most who are convicted of minor crimes are unlikely to pony up a retainer—typically $25,000 or much more—to hire a lawyer to seek justice. Nor can they expect help from the community of innocence advocates, who focus on cases where DNA can provide irrefutable evidence of innocence—usually homicides and rapes.

“Unfortunately, the Innocence Project would never take cases like these,” says Mitchell Beers, a South Florida criminal defense attorney who won an assault exoneration in 2006. About 6,000 people a year ask for help from the Innocence Project, a network of about 65 largely autonomous organizations. It has about 250 active cases at any given time, and nearly all of them focus on DNA evidence, says spokesman Paul Cates. “We are still very committed to taking cases where DNA evidence is available to prove innocence,” says Cates. “That might change at some point down the road, but the thinking is that DNA is still kind of the gold standard in proving innocence.” The Innocence Project has had a role in 325 exonerations since it was founded in 1992; just eight of them did not involve DNA cases: four home invasions, three car carjackings and one robbery.

In the alternate reality of TV cop dramas, every crime scene is scraped for biological evidence. The real world doesn’t work that way. DNA evidence is rarely collected in the ham-and-egg cases—stickups, break-ins and rip-offs—that predominate on court dockets nationwide. Biological evidence is collected in just one of five crimes, nearly all of them murders or rapes. A 2010 study for the National Institute of Justice said fewer than 10 percent cent of assaults, burglaries and robberies had physical evidence examined in crime labs, compared with 81 percent for murders.

So how vast is the trove of undiscovered wrongful convictions? No one knows for sure, because there is little empirical evidence. Zalman calls wrongful convictions “one of the most remarkably loose areas of analysis in the criminal justice field.” As Sam Gross, a University of Michigan law professor and editor of the National Registry of Exonerations, has written, “The fundamental problem with false convictions is also one of their defining features: they are hidden from view.”

But interviews by The Crime Report with more than 25 scholars, advocates, attorneys and victims suggest a growing imperative to unearth the missing exonerations and fix the systemic problems that cause them. The stakes are high. The best new research suggests that between one and five percent of all convictions across the breadth of our criminal justice system are unjust. Consider this: about one million people are convicted of serious crimes in America each year. By these estimates, as many as 50,000 of them are innocent.

Few Exonerations for Assault, Burglary

Gross’ compendium of exonerations now includes more than 1,500 examples of men and women exculpated in the United States since 1989 after convictions based upon such things as false accusations, police or prosecutorial misconduct, eyewitness misidentification, faulty forensic evidence, false confessions or a combination of causes.

Gross cautions against using the registry data to draw broader estimates of wrongful convictions. But the data is the best record we have, and it plainly points out the statistical anomalies hinting at thousands of undiscovered wrongful convictions.

About 75 percent of all known exonerations have involved a homicide or sexual assault case. Exonerations for other crimes trended upward in 2014, thanks to a cluster of 33 wrongful convictions in drug cases exposed in the Houston area. Still, just 6 percent of known exonerations were in robbery cases, even though there are four times as many robberies reported annually than the total for murder and rape.

As Gross and collaborator Barbara O’Brien wrote in 2007, “There are virtually no exonerations for the misdemeanors and nonviolent felonies that constitute the vast majority of all criminal convictions, and probably include the majority of false criminal convictions as well.”

A number of different data sets drive home the point.

Of the 1.16 million violent crimes reported in the U.S. in 2013, 62.3 percent were assaults, 29.7 percent robberies, 6.9 percent rapes and just 1.2 percent murders. In whole numbers, the country had 14,196 homicides, about 80,000 sexual assaults and 345,000 robberies in 2013, according to the FBI.

There were an estimated 724,000 aggravated assaults, the most prevalent violent crime. Well over 20 million assaults have been logged in the 25 years since the registry’s data starting point of 1989. Yet the registry includes only about 50 assault exonerations. Similarly, there are fewer than 10 known exonerations for burglary, another prevalent crime, with about 1.9 million reported in 2013.

At the end of 2013, federal and state prisons held a total of 258,600 inmates convicted of property crimes, including 131,000 burglaries, according to the U.S. Department of Justice. About 721,000 were imprisoned for violent crimes, including 186,400 for murder or manslaughter, 188,000 for robbery, 161,000 for rape and 140,000 for assault. Nearly 290,000 were locked up for narcotics. 

With more than 2 million people behind bars in America, those who claim wrongful convictions for lesser felonies have little chance of being heard above the din of lifers asking for help from innocence advocates.

Advocates focus on murder and rape cases for several good reasons.

The stakes are highest in capital cases like homicide, where the sentence can be life in prison without parole or, in 32 states, the death penalty. Those convictions often can be confirmed or overturned based upon DNA testing of stored evidence. (DNA testing confirms guilt in about half the cases the Innocence Project investigates.)

The calendar is another important factor. A successful challenge of a wrongful conviction takes an average of five to seven years, according to the Innocence Project. Those convicted of lesser felonies rarely serve prison terms that long. The average time served in state prisons is just 28 months for violent offenders, 13 months for drug crimes and 12 months for property crimes, according to the Department of Justice. The average terms are 13 years for homicide, four years for rape, 34 months for robbery, 17 months for assault, 15 months for burglary, 13 months for drug offenses, and just 11 months each for larceny, auto theft and fraud.

Someone released after serving a year for burglary is unlikely to draw retroactive attention to his or her case amid what Zalman calls the “triage” of innocence advocacy. "There are so many petitions for assistance, and the selection standards are so high, that some actually innocent prisoners will be lost in the triage," Zalman wrote in a 2012 journal article. Those convicted of “run-of-the-mill felonies” don’t stand a chance, he says.

Claudia Whitman, a longtime New England-based innocence advocate who has had a role in several exonerations, says she and her peers are inundated. "Before they reach out to me, they’ve written to a hundred people, every Innocence Project, you name it,” Whitman says. “Either they get no answer or they get a form letter saying, ‘Sorry, we can’t help you.’ Every letter I get says the same thing: ’I’ve been trying for years, and nobody will listen to me.’”

Journalists get many of the same letters, and some have led to revelatory exonerations. But most get tucked in a to-do file that never gets done. Whitman shared a 2005 letter from a Michigan inmate that shows the tenacity those who claim wrongful conviction must have to find an advocate. Whitman had suggested a few contacts. The man replied:

 Yes, I have written to Doug Tjapkes at Innocent, and was referred to Keith Findley at the Wisconsin Innoc. Proj, and was told that U. of Wisconsin do not have the resources to handle cases in Michigan. Also, I have written to Rubin “Hurricane” Carter (AIDWYC), Mr. Barry Scheck of Innocence Proj/Cardozo School of Law, Ms. Elaine Utal of Centurion Ministries Inc., Ms. Barbara Jean McAtlin of Justice Denied, Ms. Kathy Swedlow of MI Innocent Proj at Thomas Cooley Law School, Ms. Joyce A. Brown of MASS, Professor Anita Hill at Oklahoma Law School, Innocence Proj of Express Legal Service, Ms. Jennifer Linzer of Center on Wrongful Conviction/Northwestern U. School of Law, Mr. David A. Moran of Wayne State U. Law School, Professor David Protess at Medill School of Journalism, Dr. Lawrence Farwell of Brain Fingerprinting Exam Proj., U.S. Sen. Carl Levin, U.S. Attorney, President Clinton, and Governor Jennifer (Granholm) of Michigan. Unfortunately, all were unable to assist me.

‘Sheer Good Luck’

Until the age of DNA testing, there was no certain method of exposing unjust convictions. In 1932, Edwin Borchard, a Yale law professor and wrongful conviction pioneer, famously wrote that most exonerations happen as a result of “sheer good luck.” DNA testing added scientific certainty to major felony cases where blood, saliva, skin, hair or semen evidence is available. But little has changed since Borchard’s time for those convicted of lesser felonies. As often as not, luck still plays a role in the few exonerations for those crimes.

In March 1990, Robert Farnsworth Jr. was working as a manager at a Wendy’s franchise in Jackson, Mich. He dropped two bags down a bank deposit chute one night, but only one was found the next morning. Farnsworth, 29, was fired. Later, he broke down during a police interrogation and said he stole the money. He immediately recanted and said he’d been coerced.

He was charged with felony theft of the missing $2,300. He rejected a plea agreement and insisted on a jury trial. A bank employee testified it was “absolutely impossible” for the deposit to vanish, and Farnsworth was convicted based on his questionable confession. He was ordered to pay restitution and received probation and a suspended jail sentence. Farnsworth gritted his teeth and took the punishment.

Later that year, a deposit from another local business disappeared down the same deposit chute. The bank inspected the device and found three bags hung inside, including the Wendy’s deposit and a third that had not been reported missing. Farnsworth, who had gone bankrupt as a result of legal fees, was exonerated. “I told them and I told them that the deposit bag had to be in that bank, and they did not believe me,” Farnsworth told the local press. “Their attitude was that I was guilty and they were going to get me. Everybody turned their backs on me.”

“It’s so idiosyncratic how we go about exonerating someone,” says Allison Redlich, an associate professor of criminal justice at the State University of New York at Albany. “Where are the missing exonerations and how do we find them? It’s really the million dollar question. If and when we can answer that, hopefully there would be no more wrongful convictions.”

Michigan’s Gross says he suspects the rate of wrongful convictions for lesser crimes like Farnsworth’s is lower than that of capital cases. Others, including Jon Gould, an American University law professor, believe the rate is higher in minor crimes because they get little attention as they are plunged through courthouse bottlenecks.

No one doubts that robberies, assaults and burglaries are subject to the same issues of mistaken eyewitnesses, the calculated police false testimony known as “testi-lying,” incompetent defense attorneys, and crime lab malfeasance that doom defendants in major cases. “It’s foolhardy to think that errors aren’t reasonably substantial in all other crimes, both felonies and misdemeanors,” says Wayne State’s Zalman.

Conviction and incarceration statistics only hint at the potential for errors. The larger picture shows a fire hose blasting new cases into the criminal justice system every day and corresponding pressure on the courts to dispose of them quickly.

American law enforcers arrested 11.3 million people in 2013, or about 31,000 each day. More than 3 million of those arrests were for narcotics crimes, drunken driving or drunkenness. Property crimes, including burglary, larceny and theft, accounted for about 1.6 million arrests, and violent crime for nearly 500,000. Even after 20 years of a steep decline in crime, the American justice system staggers under its own weight. In the various state and local courts of California, our most populous state, 1.4 million new criminal cases were filed in 2012, according to the National Center for State Courts. That’s one new criminal case for every 26 Californians.

But criminal cases represent just a fraction of U.S. court dockets. Traffic cases are by far the greatest burden on California courts, with 5.4 million cases in 2012. The state also had 1.2 million new civil court cases filed in 2012, 400,000 domestic court cases and 110,000 juvenile court cases. Across all categories, the 2012 total was nearly 8.5 million court cases, or roughly one for every five of the state’s 38 million residents. And California’s numbers were not exceptional. New York saw 715,000 criminal cases filed in 2012, or one for every 27 residents; Alabama had 235,000 (one in 20); Ohio 825,000 (one in 14); Arizona 595,000 (one in 11), and Texas 3.3 million (one in eight). Arkansas was the per-capita leader, with 519,000 new criminal cases in 2012, or one for every six state residents. Across all types of cases, about one in three residents of the states of Arizona, Illinois, Kentucky, Michigan and Ohio had contact with the court system in 2012. Texas, with 3.3 million criminal and 8 million traffic cases, had 13.3 million total court filings in 2012, or one new case for every two of the state’s 26 million residents.

Case Clearance a Priority

Courts work hard to equalize the incoming and outgoing cases—to resolve today’s crimes and infractions to make way for tomorrow’s.

The National Center for State Courts (NCSC) recommends that 75 percent of felony cases be disposed of within 90 days and 98 percent within a year. For misdemeanors, NCSC suggests clearance of 75 percent of cases within 60 days and 98 percent in six months, and for traffic cases 75 percent in 60 days and 98 percent in 90 days. Most states achieve 100 percent clearance. Illinois, for example, had 395,829 “incoming” criminal cases and 400,260 “outgoing” cases in 2012. But some states lagged. California had an 86 percent criminal case clearance rate in 2012, with new cases outnumbering resolved cases by more than 200,000. West Virginia cleared just 76 percent, with 123,000 incoming and 94,000 outgoing cases. The lowest clearance rates were in Georgia (71 percent) and Hawaii (55 percent).

Courts use plea bargains to help equalize inboxes and outboxes. And experts say the use of enticements to wheedle admissions of guilt out of the accused makes the plea bargain process a lodestone for wrongful convictions. Ninety-seven percent of cases in federal court and about 95 percent in state court are resolved through negotiated pleas. "To a large extent this kind of horse trading determines who goes to jail and for how long,” Harvard Law School’s William Stuntz and Robert Scott wrote in an authoritative 1992 journal article. “That is what plea bargaining is. It is not some adjunct to the criminal justice system; it is the criminal justice system."

Why would someone plead guilty to a crime they didn’t commit? Low expectations is one answer. In 2013, the Sentencing Project reported that one in three black American males born today can expect to go to prison in his lifetime.

“I think you see a lot of pleas by people who have had previous contact with the criminal justice system, and maybe it didn’t turn out well for them,” says Zalman. “Maybe they’ve had a brother or an uncle who was locked up. Maybe they say, ‘I’m branded already so I’m screwed one way or the other. I’ll take the better of two bad options.’”

Many accused criminals are “lost in a whirlwind,” says Albany’s Redlich, a leading researcher on plea bargains. An indigent defendant typically meets his public defender, if he has one at all, just moments before he steps into court. Often, the defender has worked out a deal with the prosecutor: little or no jail time for a guilty plea. He recommends that the offender accept the offer, and most grab it. The New York Civil Liberties Association explored this meet ‘em-and-plead ‘em convention in a September 2014 report, “State of Injustice.” It said that one in three poor criminal defendants in Onondaga County, N.Y., never meet their appointed defender before court. In several New York counties, criminal defendants routinely appear at arraignment without an attorney. Anecdotal evidence suggests this is how business is done in many American courtrooms.

 Mark Denbeaux, a Seton Hall law professor who has studied plea bargains, suggests that those accused of routine street offenses—fistfights, shoplifting, prostitution, and even traffic infractions—are keenly susceptible to the “overwhelming temptation” to accept a plea deal to get out of jail, even if they are innocent.

Local lockups overflow with this class of offender. In 2013, American law enforcers arrested 1.1 million people for simple assault, 468,000 for disorderly conduct, 444,000 for drunkenness, 355,000 for other liquor law violations, 201,000 for vandalism, 93,000 for receiving or possessing stolen property, 56,000 for loitering or curfew violations, 49,000 for prostitution and 26,000 for vagrancy.

In a March 2013 report, the Drug Policy Alliance described a bail crisis among indigent accused petty criminals in New Jersey, where four out of 10 county jail inmates were locked up because they were unable to post even nominal cash bails. Most were blacks or Latinos, and half of were charged with nonviolent offenses. Unable to make bail, many accept the Hobson’s choice of pleading guilty to gain freedom. 

“The functional effect,” according to Alexander Shalom, senior staff attorney of the American Civil Liberties Union of New Jersey, “is that it extracts guilty pleas from innocent people on a routine basis.” (In November 2014, New Jersey voters approved bail reforms for indigent defendants, though the changes won’t take place until 2017.) These defendants see expeditiousness as more urgent than justice.

Redlich says, “You have to ask whether these people were really rational actors when they accepted the plea, or were they making a decision based on some sort of cost-benefit analysis?”

Of course, the short-term benefit comes with long-term costs for those who later claim innocence. The criminal justice system does not embrace conviction challenges, and it is particularly unyielding to those who have a change of heart after a guilty plea. The ramifications can also include impediments to employment and sentencing add-ons for those later convicted of new crimes.

Shortcuts Lead to Errors

The arm-twisting of plea bargains is the marrow of what Zalman calls our “sloppy” system of justice.

“If you look at any system that’s under a lot of pressure to produce outcomes, you get these little shortcuts and work-arounds that people create to hurry the process,” says James Doyle, a Boston attorney and former National Institute of Justice visiting fellow. “Then these shortcuts become the new normal.”

Doyle and other justice theorists are guided by the research of several scholars who work outside of crime. These include Lucian Leape and Donald Berwick, medical doctors who in 1999 began a campaign to urge their profession to act on the growing incidence of systematic errors. Atul Gawande, a Boston surgeon, writer and public health researcher, helped expose to the public the failures that afflict modern hospitals, from bacterial outbreaks to a surgical blunder that borrows a label from criminal justice: wrongful amputation. These doctors built upon the work of sociologist Diane Vaughn, who investigated the 1986 space shuttle disaster in her 1997 book, The Challenger Launch Decision: Risky Technology, Culture, and Deviance.

Vaughn blamed the shuttle explosion not on overt misconduct but on a series of bad decisions that snowballed. A phrase she used to describe that string of miscalculations has become part of the criminal justice reformer’s canon.

“It’s the most elegant statement we’ve ever seen to describe why those astronauts died, and it resonates in just about every wrongful conviction case you will find,” says C. Ronald Huff, a retired Ohio State University criminologist and early wrongful conviction scholar. “She called it ‘an incremental descent into poor judgment.’”

Vaughn’s analysis made the Challenger disaster seem inevitable—not aberrative. Boston’s Doyle calls it the “normalization of deviance.” In “Mending Justice,” a September 2014 publication that was part of the National Institute of Justice’s Sentinel Events Initiative, Doyle explained how certain events signal systemic problems in complicated bureaucracies like criminal justice.

A sentinel event in the criminal justice system warns us of threats to justice, and it calls us to act. It is a significant, unexpected negative outcome that signals a possible weakness in the system or process. Sentinel events are likely the result of compounded errors and — if properly analyzed and addressed — may provide important keys to strengthening the system and preventing future adverse events or outcomes. Sentinel errors leading to wrongful convictions have many facets, Doyle says.

“Most of the time you can’t blame it on a single cause,” he says. “When things go wrong, it’s an organizational accident. Criminal justice is this vast, fragmented, complex ecosystem, and everyone in the system feels the pressure to produce—cops, DAs, judges…And there’s fresh pressure every day. Cops have today’s cases to worry about, so they don’t have time to go back and look at the old ones for errors.”


Those implicated in an unjust outcome manage to conjure an excuse or rationalization for how and why, Doyle says. That includes bad actors like Louis Scarcella, a former New York City homicide detective under investigation for allegedly framing numerous defendants in his deep portfolio of cases from the 1980s and 1990s.

Editor’s note: For an account of recent cases associated with Scarcella, See Why Do Bad Cops Escape Punishment by Hella Winston in The Crime Report, Jan 20, 2015.

“Even in an outlandish case like Scarcella, you have to ask yourself what was it in his work environment that made this seem like a good idea to him at the time,” says Doyle. “This guy had his reasons for what he did. They all do.”

Defense Attorney Alan Dershowitz has long assailed the police lies, both petty and grand, that become foundational to prosecutions. In a newspaper commentary 20 years ago, he wrote that testi-lying “has long been an open secret among prosecutors, defense lawyers, and judges.” In the same era, Christopher Slobogin, now director of criminal justice at Vanderbilt’s Law School, published a damning article suggesting that lies are knit through our justice system. He wrote:

The pressure to lie comes at the police from all sides. Peers routinely engage in deceit, supervisors stress quotas, and the public wants criminals behind bars without having to hear too much about how they got there. The criminals themselves lie all the time, and the police naturally enough would prefer to see them incarcerated rather than out on the street two weeks after they are arrested…Police lying is not always a calculated assault on our Fourth, Fifth, and Sixth Amendment rights. For instance, at the time they engage in a search or a seizure police usually believe, in good faith, that they have the goods on the suspect. But when they truthfully explain themselves to a judge, they often find that their suspicion, based on experience and gut feeling, was an unconstitutional "hunch."

Lying almost certainly is a byproduct of the disillusionment that is pervasive among cops and others who work in criminal justice. “The first thing you’re dealing with is cynicism,” says Christine Freeman, executive director of the Middle District of Alabama Federal Defender Program. “Everyone who works in courthouses has a deep cynicism about the clientele they are dealing with. There is an assumption that they must be guilty of something; if they didn’t do this, they probably did something else.”

This presumption of guilt is a billboard factor in a justice system “defined by error,” says Bryan Stevenson, executive director of the Equal Justice Initiative, a nonprofit legal advocacy organization in Montgomery, Ala. “We have this very simplified world view, where there are good guys and there are bad guys,” Stevenson says. “A victim is good. Someone accused of committing a crime is bad. The people trying to punish him are good. The people trying to defend him are bad.”

“Part of this us-vs.-them dynamic is that we start to see accused criminals not as individuals but as entities, in a sense,” says Sara Sun Beale, a Duke University law professor who teaches criminal justice policy. “They are not really people. They are merely perps.” During her long career as a public defender, Freeman says, she has learned that her peers are as susceptible to jading as cops and prosecutors. “The thing that we’re all afraid of is that we’ve got an innocent client and our own cynicism keeps us from seeing that—from turning over that last rock that might prove it,” she says.

Stolen Stamps and an Eager FBI Agent

An FBI agent’s guilty-of-something cynicism helped paved the path to a wrongful conviction for Rachel Alaffa Jernigan.

Her life was in a spiral in 2000. At 31, she was a mother of four young children. She was also a crack addict. She had grown up in the Phoenix area in a big Latino family—10 children supported by a truck-driver father. Drug problems took hold when Jernigan was young, and she was jailed several times for narcotics and theft offenses. In September 2000, she grabbed a few panes of stamps from a rack at a local Post Office and ran out the door. A postal employee recorded the tag number of her car. She learned that she was wanted for questioning but tried to stay on the down-low.

“The smart thing would have been for me to just deal with it,” she told me. “But I didn’t want to get locked up again.”

At the same time, Special Agent Kyle Richard, the bank robbery coordinator for the FBI’s Phoenix Division, was investigating a stickup by an unusual perpetrator. On September 20, 2000, a petite Hispanic woman with a dark, blemished complexion robbed a Bank of America branch on East Guadalupe Road in Gilbert, in the Southeast Valley near Phoenix. She showed a handgun and passed a scribbled note to teller Elizabeth Chlupsa: "Don't make a big scene, give me all your money, don't give me any dye packs or tracking devices and don't press the alarm or else I will shoot." Jernigan was a diminutive Latina, and a postal inspector suggested to Agent Richard that his stamp-theft suspect might be the bank robber. Richard latched on to Jernigan as a suspect—the investigative tunnel vision that contributes to many wrongful convictions.

“That’s all he had to hear from the postal inspector,” Jernigan says. “He took my name and just ran with it.”

Agent Richard and a Gilbert police detective, Pamela Brock, created a standard six-pack photo lineup for bank clerk Chlupsa. She pointed to Rachel Jernigan’s picture. “I really feel confident,” Chlupsa said. Indicted for bank robbery, Jernigan continued to hide. “My family said, ‘Turn yourself in,’ she says. “But I knew what was going to happen. I had a record. I said, ‘They’re gonna hang me for something I didn’t do.’ And I was right.”

On the day Jernigan was indicted, October 11, while she was still on the lam, a bank in Tempe was robbed by the same Hispanic woman. She hit another bank in Chandler, Jernigan’s hometown, on October 25.

Jernigan was finally arrested on November 10. She was held in jail, unable to afford bail. As her case began moving through the system, she faced the usual carrot-and-stick offer: plead guilty and get a lenient sentence, or fight the charges and face 27 years in prison if convicted.

Jernigan chose to fight.

Agent Richard showed the six-pack photo lineup to bank employees from the two October robberies, and he secured nearly unanimous identifications of Jernigan as the perpetrator. She was charged with two additional federal counts of bank robbery—a larger Sword of Damocles to induce a guilty plea. But Rachel Jernigan would not budge. The case against Jernigan soon developed unmistakable “sentinel” problems, to use James Doyle’s term. On November 28, while Jernigan was locked up, another Gilbert bank was robbed by a short Latina. The bank was directly across the street from the Bank of America branch that Jernigan was accused of robbing on September 20. Two days later, on November 30, a robber of the same description held up a bank in Mesa, another Southeast Valley city. That robber fled in a dark Toyota SUV, the same getaway vehicle used in the October 11 robbery in Tempe.

Agent Richard now had a serial collection of five bank robberies likely committed by the same person, unless he believed that two short Latinas driving matching Toyota SUVs were independently pulling stickups in the Phoenix area. He knew Jernigan could not have committed the last two in the series because she was locked up. He also knew that the getaway vehicle connected the second and fifth robberies. The prosecutor did not share these conflicting details with Jernigan’s defense attorney, a so-called Brady violation of constitutional due process protocols for disclosure of exculpatory evidence, another wrongful conviction beacon.

She went on trial on March 21, 2001, for just one robbery, the September 20 Bank of America job in Gilbert. (The prosecutor asked to separate the two October stickups for later prosecution, apparently fearing disclosure of the messy exculpatory details.) Elizabeth Chlupsa and other bank employees identified Jernigan as the robber, and she was convicted and sentenced to 14 years by Judge Earl Carroll. It could have been worse: the sentence was lenient, a “downward departure” from the federal guidelines. Perhaps the judge sensed something amiss. Two months later, the government dropped the other charges against Jernigan.

‘Bank Robbers Get Greedy’

Jernigan was incarcerated in Florence, Ariz., a nine-prison town southeast of Phoenix. As her attorney prepared appeals, Jernigan says, she was buoyed by a fellow inmate—she prefers to identify her only as Michelle—who believed in her innocence and suggested that the true robber would eventually reveal herself.

“She told me, ‘One thing about bank robbers is that they get greedy,’” Jernigan says. “’They don’t stop until they get caught.’”

Michelle was prescient. On December 11, 2001, nine months after Jernigan’s conviction, a short Hispanic woman held up the same Gilbert bank that Jernigan had been convicted of robbing. The woman robbed teller Kathleen Golliher, who had testified against Jernigan. The robber was given money imbedded with a tracking bug. Thirty minutes later, a woman named Juanita Rodriguez-Gallegos was in police custody. She was identified by Golliher, who must have gulped at the possibility that she had been wrong about Jernigan.

Or perhaps not. Golliher stood fast, insisting that the same bank branch had been robbed 15 months apart by two different Latinas who were 5 feet tall. The mutability of memory for witnesses like Golliher is another beacon of wrongful convictions—and one that is particularly important in non-DNA cases, where convictions often stand upon the certainty of eyewitness identification. Witnesses often cling to this certainty.

Rodriguez-Gallegos was charged with the December 11 bank robbery and the two November 2000 stickups. Her case followed the criminal justice template: She took a plea deal to a gun charge, and the robbery counts were dismissed. Even then, prosecutors did not admit an error and move to open the prison gates for Jernigan. Once again, luck played a role in her exoneration.

Rodriguez-Gallegos was assigned to the same Florence prison as Jernigan. Her inmate friend Michelle learned that the new arrival was a bank robber. She told Jernigan, “It’s just a hunch, but tell your lawyer to do a search on her name.”

Her attorney, Alan Simpson, filed an appeal based on the prosecutor’s failure to disclose information about the November bank robberies by a Jernigan look-alike. She lost the appeal in both federal district court and before a three-judge panel of the 9th Circuit Court of Appeals, which voted 2-1 against her.

But the judge in that minority, a persuasive veteran named Betty Fletcher, was so certain of injustice that she apparently prevailed upon her colleagues to reconsider the case en banc, with all judges on the 9th Circuit voting—a rarity for a routine robbery conviction with no significant rules of law at issue.

In July 2007, the judges voted 13-2 to grant Jernigan a new trial. Fletcher wrote the majority opinion, which should go into a wrongful conviction attorney’s playbook. She called the eyewitness identifications of Jernigan by white bank employees “highly questionable,” and she cited influential research by psychology professor Elizabeth Loftus on “cross-race impairment” showing that members of one race have difficulty accurately describing and identifying people of another race. Fletcher also stated what should have been obvious to Agent Richard, ostensibly a bank robbery expert. She wrote, “The likelihood of two short, Hispanic female robbers with pockmarked skin holding up banks in the same area is…extremely low.”

Jernigan was confined for another seven months as prosecutors stubbornly considered a retrial. She was finally freed on February 5, 2008, after Rodriguez-Gallegos admitted guilt following a prosecutorial promise of no further incrimination.

Jernigan had served seven years and three months. In 2012, the government paid her $1 million to settle a wrongful conviction lawsuit. Today, she lives in a home in the Phoenix suburb of Avondale that she purchased with the settlement.

“I really wasn’t guilty of this crime, but it benefited me because it broke my drug addiction, made me a better person,” Jernigan says. “You can’t hold on to the past, all bitter and twisted. Sometimes you’ve got to let things go.”

She is nettled by one issue, though.

"They never even apologized to me,” she says. “I think Kyle Richard owes me an apology and that he should be a man about it and say, ‘You know, people make mistakes, and this was a mistake on my behalf.’”

I asked Jernigan whether there are other cases like hers.

“I’m sure there are a lot of people, many many people, who are in the system, who have been convicted of crimes they didn’t do,” she says. “We just don’t know about a lot of the cases because no one wants to know that this stuff is going on. But wrongful convictions happen, and I’m proof.”

Attorney Simpson, whose clients have been exonerated for both capital and lesser felonies, says victims of wrongful convictions like Jernigan’s pay a unique price. Simpson represented Ray Krone, exonerated by DNA evidence in 2002 after spending time on Arizona’s death row for the 1992 murder of a Phoenix cocktail waitress. Simpson says the relative treatment of the two after their exonerations is illustrative. Krone, imprisoned for 10 years, was treated with deference and pity after his wrongful conviction was exposed. His release made national headlines, and he got a formal apology from the State of Arizona and a standing ovation in its Statehouse. The government was eager to pay him for his trouble, with wrongful-conviction settlements of about $4.5 million. He even got a new look on TV’s “Extreme Makeovers.”

 After Jernigan’s exoneration, Simpson reached out to the Phoenix Suns to ask whether they would be willing to host the woman and her children at an NBA game. He said he hoped to “bring some joy to her life.” The team didn’t return his call.

 For Jernigan, “There was some brief media attention, but that attention span was short,” Simpson said. “And, unlike Ray Krone’s experience, no one involved in her prosecution offered an apology—not one.

 “Yet their sufferings were real and in many ways similar. Non-capital defendants suffer from the same isolation, deprivation and punishment as wrongly convicted capital defendants. But there’s no fanfare for them. They are left alone, perhaps with family, to rebuild their lives and heal the wounds of an unjustified incarceration.”

The Eyewitness Conundrum

Jernigan was a victim of cascading systemic failures, beginning with an investigator’s tunnel vision that begat eyewitness misidentifications and law enforcement malfeasance—a compounding of errors of the sort documented in Vaughn’s investigation of the space shuttle disaster.

But why did the failures happen in this particular case? Gould, the American University law professor, says two distinct types of cases tend to lead to convicted innocents.

One group is the “heater” cases, typically involving a horrific crime, a sympathetic victim and great pressure on police and prosecutors to find the perpetrator. In the rush toward resolution, the designated perp is trundled along toward a wrongful conviction, as in the prosecution of the Central Park jogger rape case in New York in 1989.

On the other hand, there are the “yawner” cases—like Jernigan’s—that get little attention.

“This is the other end of the spectrum,” Gould says. “These are seemingly simple little cases. But they often have very little evidence. You might see tunnel vision by the prosecution. Perhaps there’s just one eyewitness, and maybe not even a good eyewitness, or a questionable informant.

“There might be Brady (evidence-sharing) violations or a weak showing by the defense lawyer. Together, they are really a perfect storm of systemic failure.” Gould continues, “No one asks a lot of questions about how and why the case is going forward. It may not get much time or attention from the police. The thinking of the prosecution is, ‘You know what, I’m not sure. So let’s give it to a jury and let them make the decision.'"

"These cases,” Gould adds, “are barely hitting the prosecutor’s radar screen.”

The witness misidentifications that worked against Jernigan prove to be common in the small pool of exonerations for those falsely convicted of committing lesser felonies, especially where DNA evidence was not available. Eyewitness testimony can be dramatic and persuasive, leading to a “Perry Mason” moment as an accuser points a quivering finger at the defendant. But memory has been shown to be highly malleable, and some suggest that eyewitness finger-pointing should be viewed with the same skepticism as the lie-detector test. “The dirty little secret in all of this is how inaccurate eyewitness testimony generally is,” a California judge told me. “Having an ‘eyewitness’ is viewed by the public as a gold standard. It’s not. Witnesses often get it wrong, particularly when the perpetrator is from another ethnic group.”

The peril of eyewitness testimony is now clear. Mistaken witnesses play a role in nearly three-quarters of wrongful convictions, according to the Innocence Project. Law enforcers are keenly aware of the problem. This is the conclusion of a policy paper on eyewitnesses by the influential International Association of Chiefs of Police: “Of all investigative procedures employed by police in criminal cases, probably none is less reliable than the eyewitness identification. Erroneous identifications create more injustice and cause more suffering to innocent persons than perhaps any other aspect of police work. Proper precautions must be followed by officers if they are to use eyewitness identifications effectively and accurately."

Yet, as in Jernigan’s case, many eyewitnesses refuse to admit fallibility, even after a court ruling proves them wrong. In 2003, Justin Kirkwood, 22, was convicted of robbing a craft store in New Castle, Penn. Kirkwood had a good job, no criminal record, and no apparent motive to rob the place of $170. But two store clerks fingered him as the robber, and he was convicted and sentenced to prison despite testimony from several alibi witnesses who said he was at home at the hour of the stickup. The conviction was overturned in 2005 based upon misconduct by the prosecutor, who misled jurors about phone records confirming Kirkwood’s alibi. Even after the exoneration, one eyewitness was unbending.

"If you dismiss these charges, justice will be ignored," Nora Sbarro told a judge.

Mitchell Beers, a Palm Beach Gardens, Fla., attorney, dealt with his own cocksure prosecution witness in an assault case stemming from a traffic confrontation in South Florida. One night in March 2000, several motorcyclists argued with the occupants of a car during a road-rage confrontation on a busy Palm Beach highway. They cursed one another, and a shot was fired from the car. Mary Beth Wolter, 32, a passenger on a motorcycle operated by Robert Dziadik, was hit in the hip and seriously injured. 

Dziadik told police that the gunman was a Latino with short hair and a narrow mustache. Police were unsuccessful in identifying a suspect. Nine months after the shooting, a private investigator hired by Dziadik matched a police mugshot to a sketch based on the biker’s description of the assailant. The photo showed Vishnu Persad, 23, a native of Trinidad and Tobago who had been convicted as a teenager for auto theft and shoplifting. He also was arrested at age 20 for carrying a concealed weapon, though the charge had been dropped.

The private investigator said he was led to Persad by an anonymous tip to police who passed the name to him. By the time of the shooting, Persad was a college student who hadn’t been in trouble for several years. The mugshot linking him to the case was six years old, taken when Persad was 17. Nonetheless, he was arrested and charged with aggravated battery. At trial, Persad and three fellow students said he was studying at Florida Atlantic University at the hour of the shooting. The only compelling evidence against Persad was Dziadik’s eyewitness testimony. He was convicted and sentenced to 43 years in prison.

Persad’s family hired attorney Beers, who appealed the conviction based upon inefficient counsel. (The trial attorney had failed to point out that the private investigator who identified Persad stood to gain a $10,000 reward and that Dziadik had been drinking before the traffic confrontation.) A judge overturned the conviction in 2006, calling the evidence “tainted.” The Palm Beach County prosecutor dropped the case in 2007. But Dziadik insisted he got it right.

"I know Vishnu Persad is guilty,” Dziadik told the local press. “lf he and his family think he will walk out of court and go on with his life as if nothing happened, they are very wrong.“ Eight years after Persad’s exoneration, Beers says the case stands as an example of how one eyewitness can instigate a wrongful conviction. “It’s one of the worst witness identification cases I’ve seen in 40 years of practicing law,” says Beers. “It’s a case that just makes your hair stand on end.”

Attorneys Beers and Simpson each said they give credence to research suggesting that up to 5 percent of all convictions are bad.

Beers added, “And in cases where the primary evidence comes from an eyewitness, you can double that. My guess is that 10 percent of those cases are bad.”

Conviction Integrity Initiative

Pressure from the Innocence Project, the Center for Prosecutor Integrity, the American Bar Association and the National Association of Criminal Defense Lawyers has hastened reforms that could help head off some wrongful convictions. These include standardized videotaping of interrogations, improved protocols for lineups and more cautious use of eyewitness testimony. 

The National Institute of Justice’s ‘Sentinel Events Initiative’ now supports pilot projects in three cities—Philadelphia, Baltimore and Milwaukee—where stakeholders in the criminal justice system (prosecutors, defenders, victims’ advocates, etc.) take part in “nonblaming” exercises to identify errors that have produced miscarriages of justice in their jurisdictions.

In May 2013, the University of Pennsylvania Law School was given $15 million to endow an interdisciplinary study of wrongful convictions. Risa Vetri Ferman, district attorney of Montgomery County, in suburban Philadelphia, has agreed to use her office as an incubator to test best-practice reforms. The endowment came from a Penn alumnus who had a bone to pick with U.S. criminal justice. Frank Quattrone, a Silicon Valley investment banker, spent five years fighting a white collar crime allegation. Federal prosecutors alleged that he had obstructed justice in connection with a probe of financial rigging of initial public offerings. Quattrone was convicted of obstruction and witness tampering in 2004 and sentenced to 18 months in federal prison. He spent a not-so-small fortune on appeals and won a reversal of the conviction. The government eventually dismissed the charges.

According to the Penn Law School, the Quattrone Center for the Fair Administration of Justice will apply “evidence-based approaches” like those taking hold in medicine and education.

Quattrone declined to discuss his motivations, referring to his statement when the endowment was announced: “Our system of justice may very well be the best in the world, but with each passing day the frequency and sometimes tragic consequences of its mistakes, as well as the risk of random unfair outcomes for all Americans, are becoming better understood.”

Research suggests this is true.

In a 2014 Iowa Law Review article, American University’s Gould and a group of colleagues delineate a series of mitigating factors that they say predict wrongful convictions “with a high degree of accuracy.” They include the age and criminal record of the defendant, the punitive nature of the jurisdiction, evidence-sharing violations, forensic errors, a weak defense or prosecution, and lying or mistakes by witnesses. Gould and his colleagues recommend a series of checks, including increased funding for both indigent defense and prosecutors, and the use of experienced, senior prosecutors to make decisions on filing charges.

They say prosecutions should never be based upon a single piece of evidence, whether it is an eyewitness, a confession or forensics; that forensic evidence should be analyzed as a first step in the investigation, not last; that young defendants should have an advocate when charged with a violent crime, and that defense attorneys and police should cooperate in the development of an alibi checklist to establish best-practices protocols for a fairer investigation.

The Center for Prosecutor Integrity makes its own recommendations in its “Roadmap for Prosecutor Reform,” which calls wrongful convictions “a blight upon our nation’s criminal justice system.” And last fall, the National Academy of Sciences published a 160-page guide to the perils of eyewitnesses, “Identifying the Culprit.”

Many European countries conduct routine reviews of criminal convictions, but the trend is just beginning to take hold in the United States. Most prosecutors here fought retrospective investigations of old convictions when DNA testing became possible. Some still do. Just 16 jurisdictions in the U.S. have conviction integrity units, according to the Center for Prosecutor Integrity. Most are local or county operations in big cities, including Baltimore, Cleveland, Dallas, Denver, Detroit, Houston, New Orleans, New York, Portland and Washington, D.C. Collectively, they had uncovered fewer than 100 wrongful convictions by the end of 2014, and a third of those were from a cluster of drug exonerations in the Houston area last year based on faulty police laboratory work. Despite the Houston example, the units offer little hope for those convicted of lesser felonies since most will review only cases that have biological evidence available.

Gould told me that it is time for prosecutors to act on the growing body of evidence about flawed prosecutions. He says, “Having identified those case facts that are predictive of an erroneous conviction, the wise and just prosecutor—one who is as interested in protecting the innocent as convicting the guilty—should utilize a checklist and take extra care to more closely scrutinize cases when those questionable facts appear in a case.”

Simpson, Rachel’s Jernigan’s attorney, says the first step is for the criminal justice system to acknowledge its imperfections.

“Wrongful convictions stem from common failings in our criminal justice system,” Simpson says. “Once these failings have been recognized and the errors corrected, the exonerees need attention to help them rebuild, attention to affirm they were wrongfully convicted, and an apology from those who were a part of the wrongful conviction process.

“A sincere ‘I am sorry’ is strong medicine and reaches people deeper than some may realize.” 

The Crime Report gratefully acknowledges the support of the Fund for Investigative Journalism for this project. David J. Krajicek (@djkrajicek) is a contributing editor of The Crime Report and co-editor of Crime & Justice News. A veteran police reporter, he writes The Justice Story for the New York Daily News and was a 2014 fellow with the Fund for Investigative Journalism. He welcomes readers’ comments.

This article was originally published by The Crime Report on February 9, 2015; reprinted with permission from the author.