On the morning of July 25, 2012, Houston police detained a woman named Rosa Sade Batts in an area that was known as a hub for prostitution and the drug trade. Batts had been standing in the street, flagging down cars and “impeding lanes of traffic,” according to the Houston Police Department’s incident report, and when the two officers approached, she and two women accompanying her crossed the street in the other direction. That was enough. The cops stopped the three women for “pedestrian violations”; then they started asking questions. Was anyone carrying weapons or something illegal? Batts told them that she was not, but she added that she had flour in her purse. Baking flour? one of the officers asked. Yes, Batts answered. She liked to eat it.
Skeptical, the officer inspected Batts’ purse. He found a black plastic bag with 6.95 grams of white powder inside and a long green straw sticking up from it. The cops knew what that looked like. So they put the powder into a field-test kit and got a result that couldn’t have surprised either of them: The powder was cocaine. Batts was arrested.
The 25-year-old had a rap sheet that included prostitution and possession of a controlled substance, and because of the amount of cocaine involved and her prior convictions, she was facing the possibility of 20 years in prison. But from the outset, prosecutors made it clear that it didn’t have to come to that. Batts could plead guilty, get the minimum sentence for her offense, and forgo a trial. Two days after her arrest, she took the district attorney’s deal and began serving two years in state prison.
The prosecutors offered Batts the plea deal without ever seeing a lab report. Ten months later, when the result finally came in, it revealed something surprising. The white powder wasn’t cocaine. It wasn’t a controlled substance at all. Batts had pleaded guilty to possession of a narcotic when she was carrying nothing illegal. A week later, she was released from prison. A few months after that, the Texas Court of Criminal Appeals granted Batts habeas corpus relief, vacating her conviction.
Batts was hardly the only person in Houston who had copped a plea deal for drug possession only to find out that the facts of her case were different from what prosecutors charged. Before her, there had been 45-year-old Queen Esther Jordan, detained in 2008 after white crumbs found on her face field-tested positive for cocaine. A month after Jordan pleaded guilty, the lab results came back negative for any controlled substance. There was 33-year-old Sherri Frederick, arrested on a possession charge in 2010 after police witnessed her dropping what appeared to be cocaine from her right hand. Again, the powder field-tested positive, she pleaded guilty, and the lab reports came back negative once she was already behind bars. And there was a handful more—men and women; white, Hispanic, and African American; many with prior convictions—who were arrested, charged, and sentenced in similar circumstances, only to see their cases overturned when the lab got around to testing their alleged drugs.
Then, in 2014, the number of such conviction reversals coming out of Houston exploded. When the National Registry of Exonerations issued its annual report in January 2015, it announced that there had been a record 125 exonerations in the United States in the previous year, a 37 percent increase over 2013. That jump, the registry reported, had been “largely driven by a concentration of 33 exonerations in drug cases in Harris County, Texas.”
The Houston cases were unusual for the registry. They had been overturned not because of the efforts of a crusading defense attorney fighting an unjust system but because of an internal review by a prosecutor’s office. The defendants had not maintained their innocence during their trials and incarceration. All pleaded guilty before a grand jury even indicted them.
“Plea bargaining can coerce innocent people into accepting a sentence,” said Scott Henson, executive director of the Innocence Project of Texas. He underscored the perils of a systemic bias toward getting rapid guilty pleas. “These cases highlight the pressures on the front end for people to plead guilty and just to put it behind them.”
Such light felonies often don’t result in exonerations for the simple reason that “they usually aren’t reinvestigated,” explained Samuel Gross, a law professor at the University of Michigan who is the editor of the National Registry of Exonerations. “But in my view, the pretrial guilty pleas you see in these kinds of cases could be the most common factor to wrongful convictions across the board.” That wouldn’t be limited to Houston.
“These aren’t the sexy DNA exonerations. These aren’t the classic wrongful convictions where they’ve been screaming from the rooftops from the very beginning that they didn’t do it,” said Inger Chandler, chief of the Conviction Review Section at the Harris County D.A.’s Office, when I first spoke with her in the spring.
Chandler, 42, was the prosecutor who had discovered that drug convictions like Batts’ were not aberrations but part of a systemic problem. She was also the prosecutor in charge of cleaning up the mess.
Chandler had first become aware that something was awry in March 2014, when she received a call from Eric Dexheimer, an investigative reporter at the Austin American-Statesman. A source had alerted Dexheimer to the growing number of drug-related exonerations in Texas, and the reporter took to the website of the Criminal Court of Appeals to look for more details. Dexheimer found what appeared to be a pattern.
Men and women—white, Hispanic, and African American; many with prior convictions—were arrested, convicted, and sentenced, only to see their cases overturned when the lab got around to testing their alleged drugs.
“Harris County stood out,” Dexheimer told me. “They had most of these cases, and they seemed to be arresting people for these trace amounts of narcotics.” Something else struck him: the often lengthy gaps between the sentencing, the issuing of the lab report, and the overturning of the conviction. Dexheimer was curious if Chandler knew what was going on.
She didn’t, but she began to investigate. A picture soon emerged of a system with serious flaws at every step. First, there was an overreliance on field tests. Officers were using them as the basis for their arrests, and the D.A.’s office was using them as the basis for its plea deals. But field tests have a reputation for being imprecise, and they are not used as evidence at trial. (In studies, field tests have misidentified Jolly Ranchers, breath mints, oregano, and even air as illegal drugs.) Chandler wasn’t seeing a huge percentage of tests come back with inaccurate results, but in the nation’s fourth-largest city, even a 1.5 percent error rate could produce hundreds of wrongful arrests, which, because of the high rate of plea deals that never see trial, could lead to hundreds of wrongful convictions.
Most of the suspect convictions Chandler was now unearthing resulted in short sentences—months, not years—but pleading guilty to a drug crime can have a decades-long impact. The legacy of even a minor crime can imperil an individual’s ability to find housing, take out a student loan, and get a job. (One national report on the impact of incarceration on families found that 67 percent of former inmates were unemployed or underemployed five years after release.) Then there is the draconian legacy of the Personal Responsibility and Work Opportunity Reconciliation Act of 1996, commonly known as the welfare reform bill, which subjected all individuals convicted of a drug felony (even first-time offenders) to a lifetime ban on receiving food stamps and other federal welfare benefits. Many states reversed the provision in the late ’90s, but Texas was among eight holdouts until Sept. 1 of this year, and the law that replaced it in Texas isn’t much more progressive: It renders repeat drug-felony offenders ineligible for food stamps for life.
Reversing a wrongful drug conviction, even for defendants who have long since left prison, can have a real impact on lives. But as Chandler followed the trail of the cases, she found that once a defendant had pleaded guilty to an erroneous drug offense, the system was in no rush to correct its mistakes. In Houston, the crime lab has a practice of testing evidence even in drug cases that have been settled, but Chandler learned that the lab, overburdened and stretched thin, was deprioritizing the guilty plea cases and placing them at the back of its queue. Lab technicians would test the guilty-plea-case evidence when they had a spare moment. Sometimes that moment would come only a few days after the guilty plea was entered. Sometimes that spare moment would come months or, in a few cases, years later. “We’re constantly bombarding them with a list of cases, so I understood why they were doing that,” Chandler said.
Even the arrival of an exculpatory test result didn’t lead to swift justice. Instead, Chandler found cases where prosecutors received lab reports, then waited weeks before notifying the court. She found cases where the court, once notified, would be in no rush to appoint a defense attorney. And she found cases where the appointed defense attorneys would wait nearly a year before filing for relief. “We had clogs in truly every joint of the pipeline,” Chandler said.
Two weeks after getting the call from Dexheimer, Chandler was standing in front of Harris County District Attorney Devon Anderson, briefing her on the situation. This wasn’t the first time that a prosecutor had uncovered a string of false-positive field tests that led to bungled pre-indictment convictions. In 2013, the district attorney of Travis County—home to Austin—had found 12 similar cases and decided to end the practice of offering plea deals on the basis of field-test results. (In Austin, they called such speedily resolved cases the “rocket docket.”)
Pleading guilty before an indictment has a significant benefit for a defendant: It’s cheaper. For a first-time offender charged with possession of less than one gram of a controlled substance, state statutes set the minimum bail at $2,000—out of reach for many individuals in these cases.
In Houston, Chandler was finding that the errors on her office’s own rocket docket were far more extensive. Eventually, she would identify 425 cases from 2003 to 2015 in which a defendant had pleaded guilty to a drug crime, only for a lab report to come back contradicting the plea. In more than half of those cases, the lab returned a result of “no controlled substance.” In the remaining ones, the lab found that the defendants had pleaded guilty to the wrong drug or had copped a plea deal for a greater quantity than had been present. Such discrepancies could mean the difference between a felony and a misdemeanor, state prison or community supervision. Chandler resolved that all of the cases, even the ones that seemed like technicalities, needed to be rectified.
The process for relief was on its face simple: Chandler’s team would review the facts of the cases, then refer them to Assistant Public Defender Nicolas Hughes. Hughes would reach out to the defendants, notify them of the lab-report results, and file a writ of habeas corpus, a plea seeking relief from unjust imprisonment. Once the case made it in front of a judge, both the prosecution and the defense would ask for the conviction to be overturned.
Because the two sides agreed, in theory the process should have been simple. In reality the work was a slog. Most of the wrongfully convicted defendants had already been released, and some of them were nearly impossible to find. Hughes tracked some of them to homeless shelters; a few had died. Seeking relief didn’t require exhaustive legal preparations, but Chandler and Hughes still needed to jump through multiple hoops. First, they would go to the trial court that had convicted the defendant to ask for the judge to recommend relief. Once he or she did, Chandler and Hughes would need to present their case to the Court of Criminal Appeals. Even after the appeals court vacated the conviction, the process would continue. To check the case off the master list, Hughes would need to go back to the trial court to get the original charges dismissed. When I first spoke with Chandler, it had been a year since she had discovered the prevalence of these wrongful convictions, and she and Hughes had only managed to get a third of the way through their 425 cases. “We’re trudging along,” she said, sighing.
The morning that I visited the Harris County Criminal Justice Center in downtown Houston, a couple months after first speaking with Chandler, I waited in line for 30 minutes to enter the building. Once inside, I waited another 10 minutes before I could board an elevator. The place works in its way—I got to the sixth floor, the home of the D.A.’s office, eventually—but moving through the center is like trying to navigate a maze. There’s a sense that everyone is overworked, nothing runs quite right, and someone might just get stuck in there forever. You could see how a decade’s worth of wrongful-conviction cases might pile up before anyone realized what was going on.
In a conference room on the sixth floor, I found Chandler and Hughes. The relationship between prosecution and defense is by nature adversarial and sometimes outright hostile, but with the two lawyers essentially playing for the same team, the Houston drug cases are an exception. Nonetheless, they made for something of an odd couple, the 35-year-old Hughes playing the idealistic, crusading Mulder to Chandler’s more pragmatic Scully. (Chandler: “Nick has many causes. Most of them are noble.” Hughes: “It’s just that some things don’t sit well with me.”)
Since Dexheimer’s first article on the Houston conviction reversals was published in April 2014, Chandler and Hughes have both found themselves answering frequent questions about the cases. But one comes up more than the rest. “People ask me all the time, ‘Why do people plead guilty to something they didn’t do?’ ” Chandler said. “It comes down to that one question.”
Chandler and Hughes have two answers. Some of the Houston defendants pleaded guilty because they thought they actually did have the drugs they were charged with carrying. The other group of defendants knew that they didn’t have the drugs but pleaded guilty anyway because they thought it was their best option.
The defendants in the first category hardly make for causes célèbres. Chandler showed me the file of one wrongfully convicted man who had pleaded guilty to possessing LSD; lab tests showed that he had NBOMe, another hallucinogen, which was not illegal at the time of his arrest. He was also busted with meth and a gun—not exactly a choirboy. Did the man plead guilty because he thought that he was carrying LSD? It wouldn’t be hard to believe. Did some defendants in the “no controlled substance” cases think they were in possession of real drugs when it turned out they had been sold “turkey dope”? Chandler and Hughes had no doubt, but because exonerated defendants weren’t inclined to cop to the misdemeanor of attempted possession, there was no way to be sure how many such cases there were.
If the Houston wrongful convictions included only habitual users getting screwed over by their dealers, they’d be criminal-justice curiosities, but they probably wouldn’t be of great interest to the National Registry of Exonerations. Many of the cases, though, had facts that suggested that the defendants were not simply confused about what they had. Batts’ case seemed to fit such a pattern, as did the cases Hughes found most infuriating, arrests in which “the police just picked some dust up from a car that people shared and said, ‘Well, that looks like cocaine and you’re sitting next to it, so it’s yours.’ ” Why would people like that, railroaded from the start, plead guilty?
In Texas, 98 percent of felony drug cases settle before trial, and there’s a good reason for that: Drawn-out cases cost more for everyone and create backlogs in the system. If prosecutors can wrap up a case early, they’re going to try to do it.
Securing a pre-indictment guilty plea has been the preferred method that prosecutors nationwide employ to resolve the problem, and they’ll go to great lengths to achieve it—and not just in relatively small-time drug cases. The Innocence Project has documented many cases of innocent people pleading guilty, often to avoid a possible death sentence. “You’re wasting valuable resources when you’re putting a case on the shelf that could be disposed of,” Shannon Edmonds, director of government relations at the Texas District and County Attorneys Association, a state prosecutors group, told me. Chandler said that in cases in which a defendant was found with multiple drugs, her office would often drop some of the charges so long as the defendant would plead to one or two of them. (In one case she showed me, prosecutors dropped charges for cocaine, methamphetamine, and methylone in exchange for the defendant pleading guilty to first-degree possession of codeine and a gun violation.) Such an offer can seem too good to refuse, especially for repeat offenders.
Once such a deal was on the table, it wasn’t uncommon for defense attorneys to pressure their client to accept it, no matter what the client said about his or her guilt or innocence. Rosa Sade Batts’ lawyer, Cynthia Cline, told the Austin American-Statesman in 2014, “Sometimes, the earlier offers are the best you're going to get, so you grab it before it gets to the grand jury.” Why would a defendant like Batts want to risk an unsympathetic jury and 20 years in prison when she could guarantee a two-year sentence with a plea? If you were her lawyer, and you knew she had prior convictions and a positive field-test result for cocaine, would you tell her to take the deal?
Pleading guilty before an indictment has another benefit for a defendant: It’s cheaper. For a first-time offender charged with possession of less than one gram of a controlled substance, the Harris County bond schedule sets the minimum bail at $2,000. “There are people who can’t bond out at that level,” Chandler said. That burden falls mainly on the poor and on minority groups that are disproportionately poor, as well as on the children they might leave behind while serving time for a crime they didn't commit, or just while fighting the charge. So most defendants are left with a choice between staying in jail while they wait for lab results that may or may not exonerate them and taking the prosecution’s deal, serving a month or two, and moving on with their lives. That isn’t a calculation of justice—it’s triage. But the math is awfully clear.
On Jan. 30, 2015, Harris County District Attorney Devon Anderson announced a new policy on plea deals: Prosecutors would no longer offer them in “felony controlled-substance case[s] before the identity and amount of the controlled substance has been confirmed by a crime laboratory.” (The only exception is when a plea deal gets the defendant out of jail that day.) The policy seemed like a commonsense solution. Houston prosecutors would have to correct for any bad field tests and misidentified drugs when a defendant was awaiting the case to be adjudicated, rather than after he or she was a convict behind bars.
Still, “at the beginning everybody was up in arms about it,” Chandler said. “The defense bar didn’t like it. The judges didn’t like it—they thought their dockets were going to go crazy.” But combined with efforts to speed up the processing of lab results, the new policy doesn’t appear to have brought Houston courts to a standstill, and other Texas jurisdictions are considering similar changes.
“Harris County is a bell cow,” Edmonds told me. “That change there was significant.” It’s unlikely all jurisdictions will follow Houston—Edmonds pointed out that rural counties rely on state labs, which can take even longer to process samples—but everyone in Texas has taken heed. Earlier this year, Edmonds advised his group’s members to “keep [the Houston cases] in mind next time your judge puts the squeeze on you to plead out a case before the lab results have returned.”
The Harris County D.A.’s policy might have solved one of the fundamental issues in its cases, but as the National Registry of Exonerations noted in its report, the implications weren't limited to Houston, or to drug violations. “What about innocent defendants in non-drug cases who plead guilty to misdemeanors and comparatively light felonies—assault, shop lifting, breaking and entering—in order to avoid pretrial detention and the risk of long terms of imprisonment after trial?” the report reads. “Or innocent drug defendants who plead guilty to possession of actual illegal drugs that belonged to someone else?” For those defendants there isn’t a “cheap, reliable test for guilt or innocence,” and the pressures that helped induce the Houston defendants to plead guilty would remain in force even if every prosecutor’s office in the country followed Harris County’s lead.
While the D.A.’s solution has ended the kind of erroneous guilty pleas seen for a decade in Houston, it has brought no relief to those who have already been wrongfully convicted. Only Chandler and Hughes, plowing through the backlog of cases, can do that, and even they are of limited assistance. In September 2014, the Court of Criminal Appeals slammed the door shut on monetary compensation for nearly all the wrongfully convicted drug defendants.
“At least in Texas, the term ‘actual innocence’ applies only in circumstances where the accused did not actually commit the charged offense or any possible lesser included offenses,” the court’s unanimous majority wrote. So to obtain a ruling of “actual innocence,” a defendant would have to prove that he or she did not even try to possess a controlled substance, which would be a tall order for someone who had pleaded guilty to possessing a controlled substance.
The overturned convictions still could mean something to the defendants, even the ones already released from custody. For defendants with no criminal record other than the erroneous guilty plea, it meant they would no longer face the many consequences of being an ex-felon. For defendants who were habitual offenders, it meant that prosecutors couldn’t try to enhance future sentences based on a drug crime that never happened. But as I spoke with Chandler and Hughes, it became clear that for some of the defendants, the wrongful convictions were only one setback amid a larger set of difficulties.
This appeared to be the case with Rosa Sade Batts. When I visited Chandler in her office in Houston, she sat at her desk reading some of the case files to me. She wanted to show me the diversity of the defendants and the myriad circumstances of their arrests, and eventually she stumbled on Batts’ dossier. “Oh, gosh,” Chandler said. “That looks like a new booking photo since I last saw her.” It was. Since being released in May 2013 after her bogus cocaine conviction, Batts had been back in custody, first for criminal mischief, then on another possession charge. The long-delayed lab result had freed Batts from prison, but one overturned conviction wasn’t going to break the cycle for her and many other defendants. Bigger changes than a shift in plea-bargaining policy would be needed for that, and for many of the wrongfully convicted, it might be too late.
CORRECTION [Nov. 24, 2015]: A previously-published version of this article stated that the $2,000 minimum bond was set by state statute. TakePart regrets the error.
This article was created in association with TakePart’s parent company, Participant Media, in support of its film Kingdom of Shadows.