Nine years after Roger Keith Coleman was put to death in Virginia's electric chair, his legal case lives on, a test of whether the state may have executed the wrong man.
A young coal miner from a small Appalachian town, Coleman was convicted of the rape and murder of his sister-in-law, even though no witnesses or conclusive evidence tied him to the crime.
"I promised Roger I would do whatever I could to prove his innocence, and this is the perfect case to do it," says Jim McCloskey, a Princeton, N.J., minister who reinvestigates the cases of death row inmates he believes were wrongly convicted. "The evidence is right there, perfectly preserved."
For McCloskey and other death penalty opponents eager to capitalize on a decline in public support for executions, the question of a dead man's guilt or innocence is no moot point.
Opening a new chapter in the continuing debate over capital punishment, they have launched an effort to find an innocent person executed in the recent past.
Already, DNA testing has proved that innocent people have been sent to death row. Since 1993, 10 inmates across the country have been released from death sentences after DNA tests showed they were not guilty of the crime that condemned them.
Those mistakes have shaken confidence in the system of capital punishment, even among supporters.
Supreme Court Justice Sandra Day O'Connor, who has voted consistently over 20 years to uphold the death penalty, said earlier this month that the number of recently freed death row inmates suggests that "the system may well be allowing some innocent defendants to be executed."
That has yet to be demonstrated, but the proof may not be far away. Indeed, it may well be available in crime labs, kept in cold storage.
Death penalty foes long have suspected that innocents are scattered among the 725 people executed since the Supreme Court revived capital punishment 25 years ago. And some of them are looking to DNA to prove it. New DNA tests of old samples of blood or semen could demonstrate, once and for all, whether an executed man was innocent or guilty.
"DNA has the ability to reach back to the grave," says Richard Dieter, director of the Death Penalty Information Center in Washington. So far, there has been "no definitive proof of a fatal error [in a capital case], but there are some cases out there where it is possible it could be proven."
Dieter and others believe that showing beyond question that a state executed an innocent person could have a profound effect on the national debate concerning capital punishment.
"It could be very important. One concern about the death penalty is that it is an irrevocable act," Dieter says.
At the top of his group's list of "doubtful" executions is the case of Coleman, who was executed in 1992 for the murder of Wanda McCoy.
Perhaps not surprisingly, state authorities are none too eager to cooperate in the effort to reopen the case. Last month, Virginia officials went to court to block any new tests of the semen samples that helped convict Coleman.
A blood typing test administered shortly after his arrest had put Coleman in the 13% of the population that could have been a source of the sample found on McCoy. A disputed lab test before his execution had gone further in pointing to Coleman. A more sophisticated DNA test developed not long afterward could have shown absolutely whether Coleman was the rapist and murderer.
But Virginia authorities argued that a jury had convicted him beyond a reasonable doubt, and further lab tests appeared to confirm the result.
"There is no remaining factual issue as to Coleman's guilt," and "there simply is no important interest in retesting" the evidence of a man who was executed long ago, says Pamela Rumpz, an assistant attorney general for Virginia, in a court brief filed in June.
Lawyers for Centurion Ministries, McCloskey's nonprofit group, had sought a court order permitting a DNA test. They were joined by the Boston Globe, the Washington Post and two Virginia newspapers that had taken an interest in Coleman's case.
His pending execution in the spring of 1992 had drawn national attention. Coleman steadfastly had proclaimed his innocence, and even the Virginia judges who upheld his conviction said the case against him was "entirely circumstantial." No one saw him at McCoy's home on the night of the murder, and no fingerprints were found.
Time magazine featured him on its cover and described Coleman as someone who "might be innocent" but was due to die in the electric chair.
The original crime was swift, violent, unprovoked and inexplicable in the small Appalachian town of Grundy, Va.
On a March evening in 1981, McCoy, 19, was alone in a small rented house that sat on a hill above the fast-flowing Slate Creek outside of town. Her husband, Brad, got off work nearby at 11 p.m. and would be home a few minutes later.
Coleman, then 22, was due to work the night shift at the mine, but his work detail was called off. He stood outside the mine entrance about 10:30 p.m., chatting with a high school friend, and then hopped in his truck to drive the eight miles back down the hill toward town.
When Brad McCoy arrived home, he found his wife lying in blood on the bedroom floor, her throat slashed.
Coleman, who was married to Wanda's sister, had been convicted of an attempted rape four years earlier. Police quickly focused on him as their prime suspect.
They had little evidence except the blood test and a pubic hair found at the murder scene. A crime lab specialist testified that the hair sample was consistent with Coleman's.
A year later, he was convicted and sentenced to die.
Years later, McCloskey took an interest in Coleman's case and spent months reexamining the facts. He became convinced that two brothers who lived near the McCoys had been involved in the murder. Two local women told him that one of the brothers had bragged about killing "the woman on Slate Creek." In an odd twist, one of those women died of a drug overdose a day after telling a local TV station of what she had heard.
The Virginia authorities were unswayed, however. They bitterly resented the national spotlight put on Coleman's case, and they fought moves to reopen the murder investigation.
Virginia has carried out 82 executions since 1976, second only to Texas. But it has been stung by reversals in several high-profile cases. In 1994, Earl Washington, a mentally retarded defendant, was spared from execution by then-Gov. L. Douglas Wilder, and a DNA test later proved he was innocent of the rape and murder that had sent him to death row.
But Wilder refused to intervene in Coleman's case, and a state lawyer insisted that no further delays were warranted.
Coleman was executed as scheduled May 20, 1992.
Last month, the state succeeded in blocking the new DNA testing, at least for now.
After hearing arguments, a state judge in Grundy said he saw "no benefit" to reopening the case.
"How can investigation of the death penalty as it was implemented in 1992 be beneficial in scrutinizing the death penalty as it is carried out in 2001 when the processes are so different?" Judge K.R. Williams wrote in rejecting the request.
But McCloskey and his lawyers are not about to give up. They plan to appeal to the Virginia Supreme Court, where their prospects are uncertain, and also hope to make the debate over the testing an issue in this fall's governor's race.
"We would pay for the test. And we'd be happy to have it done in their own crime lab," McCloskey says.
Perhaps the man most puzzled by the state's refusal to allow more testing is Dr. Edward Blake, the forensic specialist in Richmond, Calif., whose tests in 1990 seemed to confirm Coleman's guilt.
Using an early type of DNA test, Blake reported that the semen sample had two markers consistent with Coleman and 3% of the population. But defense lawyers noted that a third marker was found, and it suggested that two men, not one, were involved in the crime.
Blake says he wants to use a modern DNA test on the slide but will not violate a court order to do so. But he also refused to send the sample back to authorities in Virginia.
"It's been sitting there undisturbed in the freezer for almost 12 years. It should have survived the storage well, and if so, it still has a story to tell," he says.
Blake believes that Coleman likely was guilty, but he gets angry when asked about the state's demand to return the evidence.
"The state of Virginia has no legitimate interest in undermining this inquiry. They want to seize this evidence and flush it, and I'm not going to go along with that.
"What's the problem? The guy is dead. It's obvious to me all they are worried about is the possibility--and it is a slight possibility--that this will show they carried out the execution of an innocent man."