Blood, Bullets, and Bones Read online

  Then, in 1998, Johnnie was given new hope. Illinois passed a law allowing prisoners to have physical evidence from their cases tested, if the test hadn’t been available at the time of their trials. Johnnie asked to test the pants and the victims’ fingernail scrapings for DNA. The motion was denied. The Illinois Supreme Court ruled that YT Savory had already testified that the blood on the pants was his. The court didn’t mention the fingernail scrapings. The court also said that blood evidence on the pants alone wouldn’t clear Johnnie of the crime. The testimony of the Ivys and Johnnie’s own interview weighed too heavily on the case.

  Johnnie continued to fight for the right to test DNA evidence gathered during the investigation. He was released from prison on parole in 2006, after serving almost thirty years. But he still wanted a third trial to prove his innocence. Northwestern University School of Law’s Center on Wrongful Convictions of Youth took up Johnnie’s case, and in 2013, a Peoria County circuit judge allowed testing of DNA evidence.

  Johnnie’s team tested the bloody pants, the knife found in the pants, fingernail clippings from Connie and James, semen from the rape kit, and a bloody light-switch plate from the crime scene. Unfortunately, the DNA in much of the evidence had decayed. Also, the hair evidence had been lost while in the state’s custody. However, blood left on the light switch at the crime scene still retained its DNA. That was important because the prosecution had argued that James’s blood had been transferred to the light-switch plate from the murderer’s bloody hands. James’s DNA was found in the blood, but so was a second DNA profile, matching neither of the victims, nor family members living in the home. Most importantly, it didn’t match Johnnie’s DNA. Johnnie’s attorney, Joshua Tepfer, said, “It is the killer’s DNA—and these results prove that the killer is not Johnnie Lee Savory.” The semen from the rape kit also did not match Johnnie’s DNA. Based on this evidence, Johnnie Lee Savory was pardoned by the governor of Illinois on January 12, 2015. Now a middle-aged man, he was still seeking a new trial to prove his childhood innocence.

  To date, DNA evidence has proven 329 wrongful convictions in America. At the same time, 141 real culprits were identified, who had together committed 145 additional crimes while out on the streets. DNA evidence may seem like a magic wand that erases past mistakes. But in fact, it is available in only a small percentage of cases. Even when it is, it can be difficult for a defendant to gain access to that evidence. All fifty states have statutes allowing prison inmates access to DNA evidence from their cases. But there are hurdles. If a defendant confessed—or even pled guilty as part of a plea bargain—he or she is denied access to DNA evidence. Sometimes, the DNA evidence from the case wasn’t adequately preserved. The Innocence Project is an organization that helps those convicted of crimes prove their innocence, often through DNA evidence. It recommends that the courts allow access to DNA testing in any case in which the evidence can establish innocence. It also recommends that DNA from past crime scenes be run through CODIS so that another culprit can potentially be found.

  The Innocence Project is also examining how wrongful convictions occur in the first place. In many cases, human error, or just plain treachery, led to a wrongful conviction. In a study of 325 wrongful conviction cases, 235 involved the misidentification of a suspect by an eyewitness. In 88 cases, the suspect had made a false confession—often because of police coercion. In 48 cases, incorrect or untrue information had been shared by police informants. Forensic science would seem to be an antidote for these mistakes and misdeeds. But in fact, bad forensic science was to blame in 154 of the cases studied.

  Innocence Project graph showing causes for wrongful convictions

  In some instances, the forensic science was sound but its significance overstated. For instance, if type A blood was found at the scene, that may have ruled out a suspect with type B blood, but it certainly didn’t implicate a suspect with type A blood—40 percent of the US population shares that blood type. If an expert said that the blood evidence meant that the suspect was at the crime scene, then he or she was misleading the jury.

  The same is true of hair evidence. In the 1909 murder of Germaine Bichon in France, a thick, long blond hair found at the crime scene led to a confession. In truth, the hair could have come from any number of women, and indeed, the long interrogation could have brought about a false confession. Today, experts use DNA testing in conjunction with microscopic analysis of hair. Nuclear DNA, the type typically used to match crime scene evidence to a suspect, is usually present only if the hair root is attached, and even then just some of the time. When that fails, the hair can still be tested for mitochondrial DNA.

  Whereas nuclear DNA is found within the nucleus of a cell, mitochondrial DNA is located outside the nucleus. Because nuclear DNA is the random combination of DNA from both parents, it is unique to each individual. But mitochondrial DNA, which comes from only the mother, doesn’t change from one generation to the next (except in the rare case of a mutation). Therefore, mitochondrial DNA is shared by all maternal relatives, which may include a mother and child, siblings, cousins whose mothers are sisters, and even second and third cousins whose mothers were cousins through their mothers. And it includes male siblings and cousins, who share mitochondrial DNA but do not pass it on to their children. As you can see, several people can have the same mitochondrial DNA. But this type of testing does narrow down the pool of suspects to a single family. (It would not, however, link the sample to a convict in the CODIS system, which gathers mitochondrial DNA samples only in missing persons cases.)

  Testing hair for mitochondrial DNA became routine at the FBI in 2000. Before then, many experts relied on microscopic analysis of hair alone and then gave misleading testimony about how specific such analysis could be. A study by the US Department of Justice, the FBI, the Innocence Project, and the National Association of Criminal Defense Lawyers found that in 95 percent of the 268 criminal trials reviewed, FBI hair analysts overstated the extent to which hairs could scientifically be matched. In thirty-two of those cases, the defendants were sentenced to death. Twelve hundred more cases involving hair analysis are currently under review. In some cases, there was other evidence against the defendant. However, all the defendants in those cases are being notified so that they can appeal their conviction if appropriate.

  The FBI has acknowledged its errors and cooperated with the review of its hair analysis testimony. The Bureau now uses microscopic hair analysis in conjunction with DNA testing, and has written standards for its hair experts to explain results in court without overstating the science. It is creating similar standards for nineteen other areas of forensic science. It’s a start. But legal experts say the courts should also be giving convicts recourse against bad forensic science. Currently, only two states—California and Texas—allow those convicted of a crime to appeal a case in which forensic testimony is recanted or proven invalid by scientific advances.

  Not all evidence has to be scientific, of course. Laundry marks used to solve murder cases in the early twentieth century simply linked types of marks to laundries and their customer lists. Nothing too scientific about that. But when evidence is presented as being scientific, then it should have a scientific basis. Before DNA testing was used in murder investigations, it was vigorously tested by scientists. The results of the tests were published in peer-reviewed journals, meaning that fellow scientists had to agree that the tests were valid and the results clear. Other forensic science techniques have not been subjected to the same scrutiny. A prime example of that is bite mark analysis. In several cases, bite mark analysis helped to convict suspects who were later exonerated by DNA testing of the saliva found around the bite mark.

  One such suspect was Roy Brown, who became embroiled in a murder investigation that he himself would later solve. On May 23, 1991, firefighters were called to the scene of a raging fire consuming an Auburn, New York, farmhouse. Near the fire, they found the body of Sabina Kulakowski. She had been beaten, strangled, stabbed to death, and b
itten. At first, the police investigated a man named Barry Bench. Sabina had dated his brother and still lived in the Bench family farmhouse. Barry had been to a bar the night of the murder. He had left the bar at twelve thirty a.m., and didn’t arrive home until one thirty or one forty-five a.m. When he was called to the scene of the fire the next day, he was seen walking near where the body of Sabina was later found. He told others he was “trying to find evidence or find Sabina.”4 Barry’s girlfriend told a friend she was “quite concerned when she found out that the body was found right where Barry was looking.”5 In spite of this strange fact, police soon turned their attention to Roy Brown.

  Roy Brown, exonerated for the murder of Sabina Kulakowski

  A year before the murder, Roy’s daughter had been removed from his home and placed in a residential care facility by the Cayuga County Department of Social Services. Afterward, Roy had made threatening phone calls to the department director, for which he had served jail time. He’d recently been released. Sabina was a social worker for the department. However, she hadn’t handled Roy’s daughter’s case, and there was no evidence that the two had ever met. Roy was put on trial, and an expert witness testified that the bite marks found on the body were consistent with Roy’s teeth. A defense expert countered the claim, saying that only one bite mark was complete enough to be analyzed, and that it showed two more upper teeth than Roy had. But the jury believed the prosecutor’s expert and found Roy guilty. He would later say, “I can still remember looking at the jury in my trial when they heard the scientist testify. That’s when I knew it was all over and I was going to prison, probably for the rest of my life.”6

  Roy did go to prison for a long time, but it wasn’t all over for him. He was determined to prove his innocence. First, he sought DNA testing on saliva found at the crime scene. But he was told that the saliva had been used up during earlier testing. So Roy started playing detective. His copy of the trial records had been destroyed in a fire at his stepfather’s house. From his prison cell, he wrote to the sheriff’s department requesting a new copy. Under the state’s Freedom of Information Law, the department was required to give this to Roy. The copy included four statements that Roy and his attorney say they had never seen before. All pointed to Barry Bench as a suspect.

  Roy then wrote to the courts, saying that important statements had been withheld from him, in violation of state law. For instance, Barry said he’d driven directly home from the bar, but that meant he would have passed the farmhouse. By that time, it would have been in flames. Surely he would have done something about that. The court denied Roy’s request for a new trial, saying that even if he hadn’t been provided the Barry Bench documents, the information was not important enough to overturn the conviction.

  Roy then went straight to the source. He sent a letter to Barry, saying that a DNA test would prove his guilt. He wrote, “Judges can be fooled and juries make mistakes, [but] when it comes to DNA testing there’s no mistakes. DNA is GOD’s creation and GOD makes no mistakes.”7 Five days after the letter was sent, Barry jumped in front of an Amtrak passenger train, killing himself. In 2005, the Innocence Project took up the case and discovered that saliva from Sabina’s shirt was still available for testing. It did not match Roy’s DNA. Project attorneys tracked down Barry’s daughter. She gave a sample of her DNA, and half of it matched the crime scene saliva—typical for a parent-child DNA comparison. After fifteen years in prison, Roy was exonerated in 2007.

  This is not just one man’s tragedy. According to the Innocence Project, twenty-four others were falsely convicted based on bite mark analysis and later exonerated by DNA testing. Bite mark analysis has a rather sketchy history. During the Salem Witch Trials, Reverend George Burroughs was arrested for witchcraft based on bite marks he’d supposedly left on girls. (According to the prosecutor’s theory, the reverend had bitten the girls in order to turn them into witches.) The prosecutor looked into Burroughs’s mouth and compared his teeth to marks found on the girls. Based on this evidence, Burroughs was hanged. Twenty years later, he was declared innocent—along with so many victims of the witch trials.

  Engraving of a Salem witch trial

  Such flimflam evidence could be expected of the Salem Witch Trials, where testimony about a woman seen shapeshifting into a cat was also allowed. But the same type of evidence was allowed three hundred years later, when Walter Edgar Marx was convicted of the California killing of a woman based on bite mark analysis. An appeals court upheld the 1974 conviction, saying that even though there was no scientific research to support bite mark analysis, the judge in the case had seen the bite mark comparison, and it looked good to him. This decision caused bite mark analysis to be allowed in other trials, only to be shown to be faulty later. In one case, an expert witness testified that the suspect’s teeth matched a bite mark in a bologna sandwich found at the crime scene. The suspect was convicted, but an appeals court overturned the decision when the defense learned that the other half of the bologna sandwich was found in the victim’s stomach during the autopsy, making it likely that the bite mark belonged to the victim. It’s easy to single out bite mark analysis as unsound science, but other methods described in this book have also been questioned, including fingerprint and blood pattern analysis. Could some of the convictions described in this book have been wrongful due to faulty evidence? It’s possible (and in the Cora Crippen murder, probable).

  But that’s not to say that older forensic science techniques should be thrown out the window. For one thing, they’re needed to solve cases. Shows like CSI have led some people to believe that DNA evidence can crack any case. Legal experts call this the CSI effect. Jurors are reluctant to convict in cases where no DNA evidence is available, even when there is other solid evidence to go on. But in truth, slam-dunk DNA evidence is rarely left behind at crime scenes. Think about it: criminals may shed the blood of others, but how often do they shed their own blood, or, except in cases of rape, other bodily fluids?

  That said, small samples of DNA—known as touch DNA—are often left behind when a suspect touches something. For the past ten years or so, investigators have gathered not only bodily fluids and hairs for DNA testing, but also the microscopic skin cells shed by suspects. Today’s DNA tests are so sensitive that they can create a genetic profile from the skin cells found in a single fingerprint. However, this type of evidence is controversial. A test that can analyze such a small amount of DNA can also pick up DNA from someone who was never even at the crime scene. Say I shake hands with an old friend. That friend then goes and robs a bank. My skin cells, lingering on her hand, could be transferred to the bank safe.

  University of Indianopolis scientists studied the likelihood of this scenario in 2015. Forensic scientist Cynthia M. Cale and her team had several pairs of people shake hands. Then each of them handled a clean knife. In 85 percent of the cases, DNA from the person who merely shook hands with the knife handler was found on the knife, and one-fifth of the time, the handshaker’s DNA appeared to be the primary or only DNA sample found on the knife.

  This phenomenon has had real-world implications. In 2013, a California man, Lukis Anderson, was arrested and held for several months when his DNA was found in the fingernail scrapings of a murder victim. But Lukis had a solid alibi: he was in the hospital being treated for intoxication at the time of the murder. As it turned out, the paramedics who transported Lukis to the hospital later responded to the murder scene and inadvertently transferred Lukis’s DNA to the victim. That case, and the University of Indianapolis study, show that even cutting-edge science like touch DNA has its limits.

  Old-fashioned as they may seem, footprints, fingerprints, hairs, and other physical evidence are still important to solving cases. The dawn of DNA hasn’t changed that. But exonerations based on solid DNA evidence have shown that forensic science methods need to be tested, and that the scientific limits of each discipline need to be stated clearly. In 2009, the National Academy of Sciences released a report recommending
that Congress create an independent board to establish standards for forensic evidence and techniques. The Department of Justice and the National Institute of Standards & Technology have since created the National Commission on Forensic Science, which will create uniform standards for forensic science. Their aim is to ensure that forensic science produces reliable evidence from scientifically rigorous methods.

  The US attorney general will decide whether federal and federally funded labs will follow the recommendations. States will decide the extent to which they will follow the recommendations. The standards will also be a useful tool for attorneys on both sides of cases. Prosecutors and defense attorneys alike will be able to criticize expert witnesses whose forensic science doesn’t meet the standards.

  Now is an exciting time in forensic science. Investigators are using more and better techniques to bring about justice for victims, their families, those wrongfully accused, and communities shaken by violence. Today’s forensic science is built on the work of the past. Imperfect as it may have been, it moved crime solving from darkness into light, where it is harder for criminals to hide—and for the falsely accused to be forgotten.