Skip to main content
Find a Lawyer

A Recent Exoneration Underlines a Basic Truth
As Long as We Have the Death Penalty, We Will Inevitably Execute the Innocent

By MONTY SMITH AND ALEC WALEN

Thursday, Jan. 29, 2004

As of this writing, at least 140 people convicted of first degree murder have later been exonerated by DNA evidence. At least thirteen of them were on death row at some point. The first of those thirteen was Kirk Bloodsworth.

Most Americans know that there is at least a danger that innocent people will be executed. Yet according to a recent Gallup poll, 70% of Americans still support the death penalty. Many believe that we can ensure that the innocent are never executed, if we take further measures -- provide competent defense counsel, improve police methods, and so on. (In this vein, defendants in capital cases may soon have unlimited access to DNA testing. On November 5, 2003, the Advancing Justice Through DNA Technology Act passed the House.)

But as the Bloodsworth case underlines, this faith in the perfectibility of capital punishment is misplaced. The system can be improved, but it cannot be perfected.

One of us, Monty Smith, has spent twenty years on the highly respected Baltimore County Police Force, eight of them as a detective. Smith has conducted and supervised hundreds of criminal investigations. He was personally and professionally acquainted with the detectives in the Bloodsworth case, and can vouch that they were both honest, and first-rate. Smith was also personally sure of Bloodsworth's guilt.

Last September, however, DNA evidence linked someone else to the crime for which Bloodsworth was originally convicted. Now Smith is convinced, and has written this column to convince others, that the only way to prevent the execution of innocent people is not to execute anyone.

The Facts of the Bloodsworth Case

On July 24, 1984, a nine-year-old girl named Dawn Hamilton was raped and murdered. There was some forensic evidence, in the case but it was mishandled and contaminated. Eyewitness identification was the only evidence police had.

Two young boys out fishing saw Hamilton walk into the woods with a man. Two adults saw a man fitting the same description in the neighborhood hours before the crime. A composite sketch was prepared and shown on the local news. Two anonymous callers advised that Kirk Bloodsworth looked like the composite.

Soon after the murder, Bloodsworth's wife reported him missing. He had left Baltimore on August 3 -- previously, he had worked near the scene of the crime. Detective Capel, the lead investigator on Hamilton's murder case, located him 90 miles away in Cambridge, Maryland. And indeed, Bloodsworth fit the description of Hamilton's killer.

Bloodsworth insisted he was innocent. He cooperated with Capel, answered questions, and posed for photographs. But when one of the boys who saw Hamilton walk off with a man identified Bloodsworth in a photographic array (prepared according to standard procedure), Capel obtained a warrant for his arrest.

Meanwhile, two of Bloodsworth's female associates related stories of incriminating things that Bloodsworth had said; one mentioned he'd spoken of a bloody rock. The police had not told anyone that a bloody rock had been found at the scene of the crime -- though Capel had, in their Cambridge meeting, showed Bloodsworth both underwear and a rock. (FBI experts had suggested bringing props which might cause him to react in a guilty manner if he were in fact guilty.)

Based on all of this evidence, Bloodsworth was convicted of Dawn Hamilton's rape and murder, and sentenced to death.

On appeal, Bloodsworth won a new trial, on the ground that the prosecution had withheld evidence indicating that another suspect might have been the killer. A few weeks before the second trial, evidence of yet another suspect was made available to Bloodsworth's counsel, who chose not to pursue the lead.

At the second trial, Bloodsworth was again convicted of rape and murder. This time, he was sentenced to life in prison.

In 1993, new forensic tests discovered semen on Hamilton's underpants. DNA tests proved the semen was not Bloodsworth's. As a result, Bloodworth was released from prison.

Still, prosecutors and police, Smith included, continued to believe Bloodsworth was guilty. They believed Hamilton had been molested independently of the rape/murder, and that the semen came from this unrelated incident.

In September 2003, the DNA sample recovered from Hamilton's underpants was identified as that of Kimberly Ruffner. One month after Hamilton's murder, Ruffner had been imprisoned for another rape and attempted murder. Facially, Ruffner looked extremely similar to Bloodsworth. (Granted, Ruffner was 5' 8," over four inches shorter - but that was a distinction likely lost on child witnesses to whom both men might have seemed tall enough).

The prosecutors finally apologized to Bloodsworth. Smith realized at this moment that Bloodsworth was innocent all along.

Could A Situation Like Bloodsworth's Repeat Itself, Without the Error being Caught? Absolutely.

Much of the reason Bloodsworth was convicted was sheer bad luck -- and coincidence. He bore a close facial resemblance to Ruffner, the man whose DNA is on Hamilton's underpants. He happened to work near the crime scene. He moved out of town, without telling his wife, at a time when it looked like he was fleeing. He offered up a speculation -- that there had been a bloody rock at the crime scene -- that happened to be right on the mark. Coincidences happen.

Granted, it's possible to second-guess the work of the police and prosecutors on the case. For instance, it has been argued that they should have put the new suspect who was identified before the second trial in a lineup, to see if the witnesses might pick him out rather than Bloodsworth.

But doing so might well have failed to exonerate Bloodsworth -- or, indeed, to provide any reliable evidence at all. After all, the witnesses had already seen Bloodsworth during the first trial, so they might have been inclined to re-identify him for that reason, even if wrongly. In addition, time had also passed in the interim, adding to the unreliability of any identification the witnesses might make.

It can also be argued that the witness misidentification in this case could have been avoided if the police had used a different kind of photographic identification procedure. Studies have shown that witnesses will often pick the best match when looking over a photographic array, even if no match seems quite right. In addition, if someone in the room knows who the suspect is, then witnesses may pick up subtle clues to that effect.

These problems can be reduced by showing witnesses a series of photos, one at a time, so they cannot scan them for the best match. They can also be reduced by ensuring that the police officers present do not know who the suspect is.

Reduced error is not, however, eliminated error. Using a series of photos and a double-blind procedure has been shown to reduce witness error rate from 27% to 9% . But 9% is not zero %. Errors would still be made in nearly 1 out of 10 identifications.

Moreover, it is worth emphasizing that the errors, such as there were, in this case, did not result from incompetence or corruption. They are the kinds of errors that result from using human judgment. Therefore, they cannot be avoided.

Limiting the Death Penalty's Application Will Not Save Innocents

Death penalty proponents have offered solutions to try to address the tragedy of the execution of innocent people. Some of these solutions would have saved Bloodsworth. But importantly, none of them eliminates the very real possibility that innocents will be executed.

One proposed solution is requiring DNA testing before execution can occur. Such testing eventually did help Bloodsworth. But it cannot be counted on.

Even in Bloodsworth's case, mere failure to match Bloodsworth's DNA to that recovered from Hamilton's underpants failed to convince many people that Bloodsworth was innocent. Not until there was a match to Ruffner did police and prosecutors change their mind.

Moreover, in some cases, DNA evidence may not exist. And even when DNA is found and matched to a suspect, police may make mistakes as to the reason it is there. (For instance, the police initially thought the DNA in the semen in Hamilton's underpants came from a separate molestation incident.)

Another proposed solution is not to allow executions in cases based solely on circumstantial evidence alone. But Bloodsworth's case shows why this won't work: Eyewitness identifications, such as the ones that helped convict him, are not circumstantial, but direct evidence. Yet in his case, they were wrong. Indeed, according to the Innocence Project, mistaken identifications of this sort are the leading cause of false convictions among those who have later been exonerated. They were present in 61 of the first 70 false convictions.

Yet another proposal is to allow executions only when defendants have confessed. This would have helped Bloodsworth -- who consistently protested his innocence from the moment of his arrest. But it would not have saved many other innocent people. According to the Innocence Project, 15 of the first 70 DNA exonerations for false convictions involved defendants who had falsely confessed.

Alternatively, some death penalty proponents would allow executions only when the defendant has a prior history of violent criminal activity. This would have helped Bloodsworth -- but not, for instance, Roger Coleman.

Coleman, who had already done time for rape, was executed in Virginia in 1992 for the murder of his sister-in-law. But according to John Tucker's May God Have Mercy, there is good reason to believe the rape conviction not only itself was wrong, but also led police to assume too quickly that Coleman was his sister-in-law's killer.

Another proposal for limiting the death penalty is that we allow executions only for those who kill in jail -- who cannot otherwise effectively be punished or deterred. However, the prospect of someone in jail being framed by false jailhouse testimony, or manufactured evidence, is quite real. Moreover, there are ways to incapacitate and deter those in jail -- such as secure, solitary confinement.

An "Overwhelming Evidence" Standard Still Will Not Exclude the Innocent

Finally -- and, to many, most convincingly -- some propose that we allow executions only if there is "overwhelming" evidence. They point, for instance, to the case of John Allen Muhammad, the adult member of the Beltway sniper pair, who was sentenced to death in November 2003.

Muhammad was caught sleeping in a car with a gun port in the trunk and with the gun that matched the bullets used to kill so many. And recordings of the sniper making ransom demands matched a recording of Muhammad. The evidence against him was exceptionally strong.

The problem, however, is that "overwhelming" is inherently a subjective standard. After all, some thought the evidence against Bloodsworth was overwhelming: There were four witness identifications, his apparent flight from the jurisdiction, and his apparently incriminating remarks about the murder scene, recounted by several witnesses.

Indeed, how much better can any subjective standard be than "beyond a reasonable doubt"? This standard, meant to be the most stringent we can use, still allows for a substantial number of false convictions. The Innocence Project lists 17 convictions for murder that were overturned by DNA evidence in the year 2003 alone -- and these must only be a fraction of the total.

Isn't it likely that the same juries who err in finding guilt "beyond a reasonable doubt," might also err in finding "overwhelming" evidence?

Moreover, the problem with both standards is the temptation to fit the punishment not to the defendant -- but to the heinousness of the crime. This temptation will always distort the judgment of good people and thus endanger the lives of the innocent.

Retribution is a powerful drive. For certain appalling crimes, such as those committed by the sniper Mohammad, it may be hard to accept that the person convicted of the crime should not get the death penalty. Why let him live simply because the innocent are sometimes convicted?

But there is no absolute moral imperative to execute heinous criminals. Even in those jurisdictions that have the death penalty, heinous criminals often avoid execution - they make plea bargains, prosecutors decides not to seek the death penalty, or some juror decides, for whatever reason, not to impose it.

And there is no safe way to draw a line around only those who are truly guilty. If we have the death penalty, we will misuse the death penalty.


Monty Smith is a retired police sergeant and student at the University of Baltimore. He teaches criminal law and procedure at the Community College of Baltimore County. Alec Walen teaches philosophy of law at the University of Baltimore. He is currently working on a book on Intention and Permissibility.

Was this helpful?

Copied to clipboard