The National Registry of Exonerations contains two confirmed cases of exculpatory CODIS hits emerging after a person was convicted yet going ignored by police and prosecutors for years. As a result, two men, one from Georgia and one from Tennessee, spent years wrongfully convicted, with no way of knowing about the powerful DNA evidence that proved their innocence.
In June 2000, the body of Velma Tharpe was discovered in a grass alley in Nashville. An autopsy indicated she had been sexually assaulted shortly before her murder. After a year with few leads, police came to suspect Paul Shane Garrett—a tow truck operator—in an alarming example of police tunnel-vision. Garrett consistently denied the accusations, but the detectives in the case lied that he had confessed to them in an unrecorded interview (years later, this was later proven to be false).
DNA testing of the sexual assault kit excluded Garrett as a contributor, but police found a jailhouse informant who claimed Garrett had bragged that there would be no DNA linking him to the crime because he had used a condom when assaulting the victim. Fluid found on the victim’s face and abdomen also excluded Garrett as the contributor, but investigators and prosecutors stayed focused on him. In 2003, Garrett pled guilty to a charge of involuntary manslaughter, having been unable to afford bond and feeling “worn down” after two years in the Davidson County Jail.
In 2004, the Tennessee Bureau of Investigation (TBI) informed the detective in the case and the Davidson County DA that the DNA profile from the sexual assault kit had triggered a CODIS hit to another man, Calvin Atchison, who had prior convictions for aggravated assault and domestic violence. The CODIS hit report requested a sample be taken from Atchison for further comparison, but the detective never submitted a comparison sample and the DA took no action. Garrett remained incarcerated, and was not informed of the CODIS hit that could have cleared his name.
The CODIS hit was only acted upon seven years later, in 2011. The Nashville Police’s Cold Case Unit was reviewing unsolved murders of sex workers, and found the CODIS hit report from the TBI. They obtained a DNA sample from Atchison, which confirmed he was the contributor of the DNA collected from the victim. At this point, the cold case investigators met with Garrett, who once again professed his innocence. After reviewing the evidence from Garrett’s original prosecution, they discovered the original detectives’ lies. Garrett was released from prison that same year, and was finally exonerated in 2021 thanks to the work of the Tennessee Innocence Project. Atchison was subsequently charged with the murder.
Though the CODIS hit in Garrett’s case was ultimately discovered, thanks to the Cold Case Unit’s lucky discovery, Garrett could have been released from prison seven years earlier if the CODIS hit in his case had been properly reviewed and acted upon when it first emerged. And in fact, if the Cold Case Unit had not been reviewing similar cases to the Tharpe case, Garrett’s innocence may never have come to light.
In June 2007, a convenience store was burglarized in Brunswick, Georgia, one of a string of burglaries in the small coastal city. Later that month, police arrested Michael Googe after an acquaintance, Paul Wright, told police he had seen Googe committing the burglaries. He was sentenced to ten years in prison, with eight years suspended.
In February 2009, the FBI notified the Georgia Bureau of Investigation (GBI) that blood found on the cash register at the convenience store crime scene, and that had been submitted to CODIS during the original investigation, had returned a hit to none other than Paul Wright—the very man who had accused Googe. In 2010, the GBI reported this CODIS hit to local police and prosecutors, but they took no action. Googe and his attorneys had no way of knowing this CODIS hit even existed, let alone its exculpatory contents.
Indeed, the hit only came to light in 2015 through a one-off, grant-funded project between the GBI, Georgia Innocence Project, and the Prosecuting Attorneys’ Council of Georgia. As part of that project, the GBI provided the Prosecuting Attorneys’ Council with all 2,700 CODIS hits they received from 2007–2014, including Googe’s. The Prosecuting Attorneys’ Council reviewed all these hits and provided those where a person had been convicted of a crime, but the hit pointed to someone else, to the Georgia Innocence Project. The exculpatory nature of Googe’s CODIS hit was so plain that the Prosecuting Attorneys’ Council immediately took action, and he was swiftly exonerated.
Had Googe’s CODIS hit not occurred within that seven-year window subject to the grant project, though, it would likely never have come to light, and Googe’s wrongful conviction would likely still stand. To date, no other state has conducted the sort of CODIS hit review that Georgia did through its grant-funded project, and even in Georgia there has been no concerted review of potentially exculpatory CODIS hits since 2014.
The Accurate Justice Project was founded to address the exact failure that occurred in these cases: CODIS hits that emerge after an innocent person has been wrongfully convicted, both proving the person’s innocence and identifying the true perpetrator, yet never getting investigated or disclosed due to inadequate safeguards and procedures.
Yet, by its very nature as an issue of non-disclosure, finding confirmed cases of this failure occurring is extremely difficult. By law, the police laboratories that administer the CODIS program can only provide CODIS hit reports to police and prosecutors. As a result, when police and prosecutors fail to investigate or disclose a CODIS hit, there is normally no way to learn of its existence.
In both these cases, the improperly ignored CODIS hits only came to light years later due to extraordinary circumstances—in Georgia, a one-of-a-kind grant project, and in Tennessee, an unusually lucky investigation by a committed cold case unit. The discovery of these two cases is significant not only because it proves that this problem of exculpatory CODIS hits being ignored is very real (and with devastating personal consequences), but also because it suggests there are many more such cases that—without similarly extraordinary action—will never be known.
This is why it is essential to adopt policies to guarantee that CODIS hits are consistently and reliably reviewed, disclosed, and acted upon. The Accurate Justice Project is working with police and prosecutors to get model policies adopted to do just this.