"You can escape from prison, but how do you escape from a convincing story?”
Errol Morris, A Wilderness of Error
The murder of Catherine Fuller rocked the relative calm of 1984 Washington, D.C. Fuller, a 49-year-old wife and mother, was sodomized with a pipe, and kicked and beaten to death for $40 and the cheap jewelry she wore. It was perhaps the most savage and senseless killing in District history. That’s what the police said, anyway.
Detectives quickly called it a gang attack. By the time their investigation ended they had arrested 17 people for the crime, all alleged members of the so-called 8th and H Street Crew. After a trial in late 1985, eight young men were found guilty of the murder and sentenced to life in prison, despite their assertions of innocence. The killing had “consumed the community,” in the words of a Washington Post story. People wanted to put it behind them.
Once the appeals were denied in 1988, the case seemed to be closed forever. But it wasn’t. On March 29, the U.S. Supreme Court will hear arguments about whether the government’s misdeeds require the reversal of all the convictions.
How — and why — does a case come back to court after three decades?
Around 4:30 p.m. on a drizzly Monday afternoon, Oct. 1, 1984, Fuller left her brick row home on K Street in Northeast D.C. She was headed to Family Liquors, a short five-block walk away. Barely 90 minutes later, a street vendor found her 99-pound body on the cement floor of an empty garage, off a narrow, trash-strewn alley behind the 800 block of H Street NE.
The two lead detectives assigned to the case — Patrick McGinnis and Ruben Sanchez-Serrano — faced a steep challenge. Neither man had spent much time in the neighborhood where the crime took place, and neither was skilled at working the street. Several searches of the scene — the garage and the alley around it — yielded no physical evidence that pointed to any likely killers. The bystanders who gathered that night had no information, just questions.
But within hours the detectives had crafted a theory of the crime, based on an anonymous phone tip. Members of the 8th and H Crew, hanging out in the nearby park, had seen Fuller walk by on her way home. They pushed her into the alley to rob her. When she resisted, they dragged her into the garage, where they attacked her “like sharks on a feeding frenzy,” as Sanchez later put it. Then, as a final, awful indignity, before they left her to die, two of them held her legs open while another rammed a pole 11 inches up her rectum.
It was a riveting, unforgettable narrative. But hard evidence to support it proved to be elusive. The killing occurred during rush hour on a Monday, H Street was a busy commercial strip, and the two side streets had row houses that backed on the alley. “Because of the neighborhood as I knew it,” McGinnis said, “someone had to see the lady go back in the alley.” Yet hundreds of hours of canvassing, and dozens of interviews, turned up no eyewitnesses who had seen anything related to a gang attack.
Nevertheless, the two detectives persisted. They believed in themselves, and in their theory. They believed that 20 to 25 young people had been in the alley during the attack, either taking part or cheering the others on. And finally, at the end of November, they got the break they had doggedly pursued. After a period of intense interrogation, and some initial denials, a 19-year old named Calvin Alston said he had seen the attack, and implicated 13 other men. Although Alston soon recanted, McGinnis and Sanchez got confessions from two more teens, and by the summer of 1985, 17 people were in custody for Fuller’s murder.
As the horrible details of the crime became public, the prosecution wove them into the narrative they had constructed. The number of suspects climbed, and the gang story captured the public, the media, and even the defense lawyers as well. It was the only version of the killing that anyone heard. It became the accepted truth of what had happened, long before any evidence was presented in a court of law.
One of those arrested for the murder was Christopher Turner. Turner and a few of his friends had gone to a midnight showing of “Beverly Hills Cop” on Saturday, Dec. 8, 1984. When Turner got home in the early hours, an old girlfriend was waiting to see him. She ended up staying.
Just after 7 a.m., he woke to the sound of his younger brother yelling, “Chrissy, the police are here.” They burst into his bedroom with guns drawn. As quick as he could stand, he was in handcuffs. His grandmother had opened the front door so the officers wouldn’t break it down. When he was led outside, Turner saw several marked cars, and more police. A helicopter was circling overhead. He says he had no idea what was happening.
Turner had just turned 19. He was a graduate of Coolidge High School, planning to enlist in the Air Force. Before that day he had never been arrested, or charged with a crime, or spent a minute in custody. Suddenly, he faced a murder charge, and the possibility of life in prison.
At the homicide office, Turner told the police that throughout the day of the killing he was at the home of his best friend, Kelvin “Hollywood” Smith. Smith, who turned himself in the next day, told them the same thing. Both said they remembered October 1 because a girl had called Smith that night with the news of a murder in the neighborhood. And the next day they learned that Fuller, whom they knew from the area, was the victim.
All the accused offered an alibi defense: “I wasn’t there.” But with one exception, they had been arrested many weeks after the killing. They struggled to remember exactly where they had been, and what they had been doing, at the time of the crime. Like Turner and Smith, they lived in that neighborhood and went to school or the park or the arcade or the carry-outs there every day. It would have taken only a few minutes to be involved.
Because Turner had no record, the prosecutor offered him a sweet bargain: plead to a minor offense, testify for the government, and get a 2 to 6-year sentence. With good behavior he would have been out just after his 21st birthday. Most of the accused were offered deals as well.
But Turner wouldn’t even consider a plea. He insisted he was innocent. No matter what the offer was, he told his lawyer, he would never admit to a crime he had taken no part in.
At the trial, the government had no physical evidence linking any of the defendants to the crime — no DNA, no hairs or fibers, no fingerprints or shoe prints. There were no unrelated eyewitnesses. The prosecution’s case, in large part, was their mesmerizing story and the confessions of two teenagers, who traded their testimony for leniency. Although their accounts often differed, and were rife with mistakes and contradictions, both snitches told a tale of twisted, thoughtless evil that ended with a bone-chilling atrocity.
The defense lawyers had no story to tell. None of them challenged the prosecution’s narrative, or questioned whether Fuller had, in fact, been murdered by a group of young people. As Jerry Goren, the lead prosecutor, later told a Post reporter, “Once you say, ‘I wasn’t there,’ you can’t argue about what happened.”
The alibi witnesses who testified believed in the innocence of their family member or friend. But their memories about the day of the murder — now more than a year earlier — were not sharp. There were gaps and inconsistencies in the details they offered. In the end, the jurors had little confidence in what they said. And none of the defendants could account for every minute of that fateful Monday.
After a five-week trial and nine days of deliberations, eight young men, including Turner and Hollywood Smith, were convicted of first-degree murder.
Describing his work for O.J. Simpson, famed lawyer Johnnie Cochran explained a truth that all trial lawyers know: “Evidence doesn’t win the day,” he said. “We’re here to tell a story. Our job is to tell that story better than the other side tells theirs.”
In this case, the government had a superb, compelling story. Through telling and re-telling, with help from the media, it had become the only story of the crime. The accused had no counter-narrative of innocence, no alternate version of events. So the prosecution’s story carried the day. For many court watchers, the biggest surprise was how long it took the jury to find most of the defendants guilty.
The defense lawyers didn’t know that there was another possible narrative of Fuller’s killing: that the crime had been committed by only one or two assailants. It was a narrative that could have been as compelling as the official version of events, but also exonerated every one of the accused. The lawyers didn’t know that because the prosecution had chosen to hide the information.
There were at least three key pieces of evidence that pointed to a very different scenario.
First, a few weeks after the crime, a woman named Ammie Davis told police that on Oct. 1 she had gone into the alley behind H Street to shoot up some heroin with a girlfriend. In the alley, Davis had seen a man she knew, James Blue, “beat the fuck” out of Fuller “for just a few dollars.” The officer wrote a report about her statement, but it was somehow “lost in the shuffle,” as the prosecutor later put it.
When the report surfaced some 10 months later, in August 1985, the government’s theory was set and the trial was fast approaching. Goren interviewed Davis, but decided she was not credible, and little investigation was done regarding her story. Three weeks before the trial began Blue shot Davis. He was convicted for her murder, but died in prison in 1993, before Davis’ story came to light.
Second, the street vendor who found Fuller’s body, 19-year-old William Freeman, had waited by the garage with a woman friend to direct arriving officers to the right location. During the wait, both saw two young men next to the garage, acting suspiciously. As the first officer pulled into the alley, Freeman heard one of the men say “don’t run.” Then they both fled. As they ran, one of the men seemed to be hiding an object under his coat.
Several months later, both Freeman and the woman looked at some photos and identified the runners as James McMillan and his associate Gerald Merkerson. McMillan was the one with something under his coat. Later, two other independent witnesses also put McMillan in the alley around the time of the killing. He was living in the 900 block of 8th Street NE, in a house that backed onto the alley just steps from the garage where Fuller’s body was found.
Third, a man named Willie Luchie had told the detectives that on the night of Fuller’s murder, he and a few friends walked through the alley around 5:30 p.m.—the most probable time of the killing. As they passed the garage where her body was found, some of them heard a low moan. Luchie turned around to look. But both doors were closed, and they were headed to the liquor store, so no one stopped to investigate.
Luchie also said that he did not see anyone else in the alley; there were no groups of young people coming or going or hanging out there. Another member of the group confirmed Luchie’s story to the police.
All this evidence — Davis’s report, the four identifications of McMillan, and Luchie’s story — was withheld from the defense.
These facts would have stayed hidden without the work of a Post reporter named Patrice Gaines. She had come to the newspaper in 1985, shortly before the Fuller trial, and done some supporting work on the case. She was in court when the first guilty verdicts were announced, and they left a deep impression on her.
Gaines couldn’t shake the feeling that there was something wrong about the convictions. In the mid-1990s, after getting a letter from Chris Turner, she persuaded her editors at the Post to let her take a look back at the case. She and a colleague dug into the old files, and interviewed many of the participants. The more Gaines learned, the more certain she became that the defendants were innocent. But she couldn’t find conclusive proof to convince her superiors. In the end, the most she could do was write a story about her personal journey with the case, detailing her doubts about the outcome.
It turned out to be enough. Gaines sent a copy of her article to the Mid-Atlantic Innocence Project (MAIP), and the organization agreed to take the case. MAIP recruited some major law firms, and began to assemble the relevant materials. For the first time, the convicted men had competent counsel who believed in their innocence. They also had deep pockets. They paid investigators to track down witnesses, and made repeated requests for files and reports from the police and prosecution.
MAIP and the new lawyers turned up other new evidence that undermined the gang theory. An expert in forensic pathology said Fuller’s injuries, though extensive, were more consistent with a small number of assailants than with 10 or more. A crime reconstruction expert said the scene in the garage pointed toward just one or two attackers. Both cooperating witnesses recanted their testimony in sworn affidavits, and said they had lied to save themselves.
The government slowly turned over more and more documents—some voluntarily, some only after a court order. Among those documents was the evidence that the prosecution had withheld. The lawyers for the accused believed that by hiding this information. the government had violated a legal requirement known as the Brady rule.
In 1958, John Brady was charged with first-degree murder for the strangling death of William Brooks in Anne Arundel County, Md. According to the police, Brady and an accomplice, Charles Boblit, planned to steal money and a car from Brooks. They put a log across his driveway. When Brooks stopped to move the log, they took his car, carried him into some nearby woods, and choked him with Boblit’s shirt. They were tried separately, with Brady first.
At his trial, Brady admitted he took part in the crime, but said Boblit had actually killed Brooks. In his closing, Brady’s attorney conceded that his client was guilty. But he argued that, given Brady’s secondary role in the murder, he should be spared the death penalty. The jury disagreed, finding him guilty and recommending a death sentence, which was imposed.
Before the trial, Brady’s counsel had asked the prosecutor for copies of any statements Boblit made to the police. Several were disclosed. But one in particular was withheld: The one in which Boblit acknowledged that he had, in fact, done the killing.
After Brady was convicted and sentenced to death, his lawyer learned of the additional statement by Boblit. He filed a motion asking for a new trial, based on this newly discovered evidence that had been withheld by the prosecutor. The case eventually reached the U.S. Supreme Court, which agreed that the suppression of this evidence had denied Brady “due process of law” guaranteed in the Constitution.
The heart of the court’s decision, known since as the Brady rule, is that a defendant’s right to a fair trial is violated if the prosecution withholds evidence that is “favorable to an accused” and that is “material either to guilt or punishment.” Said the court: “Society wins not only when the guilty are convicted but when criminal trials are fair; our system of the administration of justice suffers when any accused is treated unfairly.”
Despite the Brady rule, the government had not only concealed some important pieces of evidence in the Catherine Fuller murder, but also fought to keep the identifications of McMillan secret. At the trial, the defense lawyers specifically asked the prosecutor if Freeman had ever identified the two young men by the garage. Goren argued strenuously that he was not required to disclose that information. He told the judge—incorrectly, as he would later admit—that the two were in the alley “one and a half hours after” Fuller was killed. And so, he said in open court, “it didn’t seem to me that that was Brady in any way.” He would not even confirm whether the prosecution had done any ID procedures with Freeman.
After some short arguments, the judge was not sure that Goren was entitled to withhold the information. But he did not want to delay the trial. He told the defense lawyers “we’ll take that up later.” They never did.
When this hidden information finally came to light, Goren offered a further defense of his actions. He said he did not turn over Davis’s statement because he didn’t believe she was credible. Her information—that Blue committed the murder alone—didn’t match his belief that a group had done the crime. He said he dismissed Luchie’s story because it would only be relevant if “there was a single person in the garage with her.”
Both of those statements were highly significant. The lead prosecutor was saying that because he was sure his theory was right — that the killing was a gang attack — whenever the police found evidence that didn’t support that narrative, or that suggested a different scenario, he decided it was irrelevant, and felt free to bury it.
Goren’s actions illustrate the central weakness in the Brady rule. The jury — not the judge, not the prosecutor — should decide the facts of a case, and the credibility of witnesses. As the Maryland Court of Appeals wrote in its Brady decision: “We cannot put ourselves in the place of the jury and assume what their views would have been as to whether it did or did not matter whether it was Brady’s hands or Boblit’s hands that twisted the shirt about the victim’s neck.”
The evidence withheld by the prosecution in this case was “favorable to the defense.” So the primary issue in dispute, as with most Brady claims, was whether that evidence was “material.” Court rulings have said evidence is material if “there is any reasonable likelihood it could have affected the judgment of the jury,” or if the nondisclosure “undermines confidence in the outcome of the trial.”
The lawyers representing the convicted men — now the appellants — argue that the hidden evidence is material primarily because it would have provided a strong counter-narrative to the government’s story. This claim, they say, is particularly strong when the McMillan and Luchie evidence is viewed together.
McMillan would have been a very likely suspect. He had quick, easy access to the alley from his home. He was on the crime scene, behaving suspiciously, around the time of the killing. He ran when the police came, holding something beneath his coat. The object used to sodomize Fuller was never recovered.
Despite his youth, McMillan had a record of serious violence against women. One young woman he dated briefly said later that “all he wanted was oral and anal sex.” He was also a loner, a newcomer to the neighborhood, who did his crimes solo or with a single accomplice.
Within two weeks of Fuller’s murder, McMillan demonstrated his savage streak by committing two vicious daytime robberies of women who were walking alone not far from 8th and H NE. Both were punched multiple times in the face. He later pled guilty to both robberies, and on March 15, 1985, as the Fuller case was moving forward, he was sentenced to 8 to 24 years in prison.
As for Luchie’s story, what he and his friends heard in the alley was almost certainly the attack on Fuller in progress. The time is right. But the garage is too small for a gang of 20 or 25 young people to be standing around inside kicking and beating Fuller. And Luchie and his group saw no one else during their walk through the alley.
If this information had been disclosed prior to trial, the appellants say, they could have argued that it was McMillan, with Merkerson’s help, who had sodomized and killed Fuller. The evidence for that story is substantial—McMillan’s violent past, his sexual tastes, his proximity to the scene, his suspicious flight, the missing weapon. It would also have explained one of the enduring mysteries about the case: why no one living in the neighborhood, and no passers-by, had seen anything related to a gang attack. It was because there hadn’t been one.
In sum, the lawyers said, this evidence would have led to a different outcome at the trial.
The government, of course, disagreed. They conceded that some potentially exculpatory evidence was hidden from the defense, and that under “prevailing standards” today the information would be disclosed. But their bedrock claim was that because the evidence against the defendants — primarily the testimony of the two confessors — was “overwhelming,” none of the withheld evidence was material. None of it would have made a difference in the result of the trial. Any errors were, in legal language, “harmless.”
The prosecution’s position was bolstered by the fact that recantations — however credible they may seem—are given short shrift in our legal system. One reason is that the law craves finality, and a recantation wreaks havoc on a previously decided matter. Another is that any witness who recants has shown a willingness to lie under oath.
And proving materiality in this case was particularly difficult. All the Brady evidence at issue supported a scenario of one or two assailants. But the only story of the crime for 30-plus years had been of a group attack involving at least 20 young people. Anyone who believed the gang story was factual would have to say that none of the hidden evidence raises doubts about the results of the trial.
So to have any chance to prevail, the lawyers for the appellants needed to wipe the slate clean; to move the clock back to the beginning, before there was any theory. They needed to convince the judges to look at the entire case with fresh eyes, and imagine what jurors might have done if they had heard both a group scenario and a one- or two-person scenario.
So far the appellants have failed. Both lower courts ruled for the government. Both discounted the recantations, and did what the trial prosecutor did: evaluated the evidence in light of a belief that the crime had to be a gang attack.
After a 2012 hearing in D.C. Superior Court, Judge Frederick Weisberg denied their motion for new trial. In the judge’s view, for the hidden evidence to be material, McMillan and Merkerson would have to have done the crime by themselves. And “that possibility,” he said, “flies in the face of all the evidence.” What he meant, of course, was the prosecution’s trial evidence. The appellants had presented him with a sizeable body of new information — recanted stories, expert testimony, hidden facts and witnesses — showing that the crime was most likely the work of one or two assailants. But he could not imagine any scenario other than a gang assault, no matter what the new evidence showed.
The result was the same in the D.C. Court of Appeals (DCCA). While the judges hearing the case were troubled by the prosecution’s actions, they ruled that the Brady claims “fail because appellants have not shown a reasonable probability that the outcome of the trial would have been different had the government disclosed the withheld evidence in a timely fashion.”
The judges agreed that the withheld evidence went directly to the key point of disagreement: “The basic structure of how the crime occurred.” But they then carried this point to its extreme conclusion:
This makes the burden on appellants to show materiality quite difficult to overcome, because it requires a reasonable probability that the withheld evidence (in its entirety, and however appellants would have developed it) would have led the jury to doubt virtually everything that the government’s eyewitnesses said about the crime.” [Emphasis in original.]
A bar that high is nearly impossible to clear.
In their briefs to the high court, the key issue raised by the appellants was that the DCCA decision had misapplied current law, and set the materiality standard far too high. Friend of the court briefs also argued that in recent years prosecutors have routinely violated Brady without any consequences, and that the Court should put a stop to that trend.
Because the appellants lost the first two rounds, it was a surprise to most observers that the Supreme Court agreed to hear the case. The issue for the justices to decide will be the same as it was in the lower courts: whether the prosecution’s sins were material. The government will argue that even if the hidden information had been disclosed, it would have changed nothing. The appellants will argue that it could have changed everything. Under the law, they need not prove that disclosure would have led to different verdicts, only that non-disclosure undermines confidence in the results.
Looking back, it’s easy to understand why prosecutor Goren — confronted with some difficult choices — chose to err on the side of secrecy rather than the side of full disclosure. He was firmly behind the government’s theory. He believed the young people who had been arrested had viciously murdered Fuller; that they deserved to be convicted and punished. But he didn’t have a lot of proof. He later told a Post reporter that “we were really struggling to find the evidence we needed to make the case. …It was a case that easily could have gone the other way.”
Goren knew that disclosing this information might well derail the entire prosecution. If McMillan had sodomized and killed Fuller, the 17 people he had accused were innocent. If Luchie and his friends heard the assault in progress as they passed the closed garage, it could not have involved a gang. After all the time and effort, and all the publicity, it was impossible to even think that he might be wrong. The cost was too high.
So Goren made himself the sole judge of the credibility and relevance of this information. He didn’t want to believe it could be true, so he didn’t want the defense to hear it.
We know, of course, what did happen. Fuller was horribly murdered, and her family lost a wife and mother and sister. We don’t know if they got justice.
We know that eight men — who all swear they are innocent — have so far spent a total of 232 years behind bars. One died of an aneurysm in 1999. Chris Turner was paroled in 2010 after serving more than 25 years. The other six remain in prison.
After some early struggles, they all have had good records. Like Turner, in the decades since the crime not one of them has ever said he had any part in the killing. That refusal has kept them from any chance at early parole. Even now, after more than 32 years inside, they will not trade their integrity for their freedom.
Because of his exemplary prison conduct, Turner is a rare exception: someone who made parole despite maintaining his innocence. But he spent 25 years of his prime behind bars. He lost his dreams of a career in the Air Force, of a home and family. His two sons grew up without a father in their lives. His beloved grandmother died before the trial, and he believes his arrest hastened her death.
As for James McMillan, he later provided one more terrible reason why he should have been a suspect in Fuller’s murder.
After serving almost eight years for his two brutal robberies in the days after her killing, he was released to a halfway house in D.C. on July 23, 1992. Less than two months later, on Sept. 15, 22-year-old Abbey McClosky was walking home around 8 p.m. when she was dragged into the alley behind the 500 block of 8th Street NE and brutally assaulted.
Shortly afterward, two young men walked into the alley because their car was parked there. As they started to get in, a man stood up from behind the car. He was pulling up his pants with one hand, and holding something in his other hand. He mumbled what sounded like “Excuse me,” and then quickly left the alley. When the two men checked behind the car, they saw legs sticking out. Thinking the person might be dangerous, they went back to their apartment and called 911. Both men later identified McMillan as the person they had seen in the alley.
McClosky died three days later without ever regaining consciousness. An autopsy showed the cause of death to be “blunt force trauma to the head and torso.” Specifically, the skin around the anus had “radiating tears,” and there were rips and bruises extending up into the rectal mucosa which “clearly indicated sodomy or forced penetration.”
There were several obvious parallels between Fuller’s murder and that of McClosky. Both were small women, attacked in alleys near 8th and H NE. Both were horribly beaten, with injuries more like a car crash than a typical assault. And in both cases, the assailant had focused his fury on the victim’s anal areas.
In 1993, McMillan was convicted of McClosky’s murder. He was sentenced to life in prison without the possibility of parole.
A note on sources: Details of Catherine Fuller’s murder, as well as the murder of Abby McCloskey, come from police documents about those cases, including reports, interview notes and transcripts, and case resumes and summaries. Quotes from the detectives and prosecutor are from transcripts of interviews conducted by Washington Post reporters in 1996 and 1997. Information on newly discovered evidence and post-conviction legal matters is from briefs filed by the parties for the new trial hearing in D.C. Superior Court in 2012, the arguments in the D.C. Court of Appeals in 2014, and the upcoming argument in the U.S. Supreme Court, plus transcripts and reported decisions from those proceedings. The information about Chris Turner comes from numerous interviews conducted with him from 2010 through 2017.
Thomas L. Dybdahl, who has degrees in theology, journalism and law, retired in 2013 after 13 years as a staff attorney at the Public Defender Service for the District of Columbia, where he worked in both the trial and appellate divisions, and tried 25 homicide cases. He is writing a book on the Catherine Fuller murder case.
