Last month’s arrest of NorCal Rapist suspect, Roy Charles Waller, sent shockwaves across the Cal campus and the state. Waller, after all, was a longtime employee of UC Berkeley’s Office of Environment, Health and Safety, and his capture resulted from a new forensic tool that promises to solve many cold cases: open-source genealogical databases.
The technique gained widespread publicity in July with the capture of Golden State Killer suspect, Joseph James DeAngelo. With DeAngelo—as with Waller—police uploaded information on DNA to GEDmatch, an open source site that allows users to find long-lost relatives. Hits on some of DeAngelo’s relatives ultimately led to the arrest of the 72-year-old suspect at his Citrus Heights home.
“The problem is that as time passes, it becomes increasingly difficult to mount a meaningful defense. Memories fade. Witnesses may die,” says Andrea Roth.
Waller’s case also involved the issuance of a so-called John Doe warrant—a complaint filed by the Contra Costa County District Attorney’s office based on DNA obtained from several rapes and assaults. A John Doe complaint allows prosecutors to avoid statute of limitations expirations in an unsolved crime, given that a crime has been confirmed and a complaint issued for a suspect, even if that suspect is only known by their genetic signature. In Waller’s case, the warrant was filed in 2006, the same year that marked his last known assaults.
Combined with genetic profiling, John Doe warrants have the potential to resolve an untold number of heinous crimes. At the same time, civil liberty activists worry that these procedures put basic constitutional rights—privacy and the presumption of innocence—at risk. The question remains whether the benefits of new forensic methods are worth the potential cost.
The California Supreme Court confirmed the legitimacy of John Doe warrants in a 2010 ruling. They are legal in many other states—but not all. Nor are they allowed in federal courts. The rationale against them is the right of freedom from prosecution for major crimes without a grand jury indictment, a protection assured by the 5th Amendment. So, why aren’t they verboten in all states?
“The U.S. Supreme Court has made it clear that the right to indictment is not so fundamental [as some other constitutional rights],” says Andrea Roth, a Berkeley law professor and the faculty director for the university’s Center for Law and Technology. “The right to counsel, to a jury trial—those have all been confirmed as universal by the courts, and all states must conform in guaranteeing and protecting them. The right to indictment isn’t considered comparable.”
In the case of rape, says Jon Simon, law enforcement has largely failed. Victims often are unwilling to come forward because of trauma or concerns that they won’t be believed.
Roth says she doesn’t think John Doe warrants pose a major threat to the right to privacy, “but I do have great concerns about their potential to undermine the right to rebut witnesses and compel favorable testimony [for defendants]. And I’m also alarmed about their use to get around the statute of limitations. We already have an exemption to the statute of limitations for homicide. Now these warrants essentially do the same thing for sexual assault and possibly property crimes. The problem is that as time passes, it becomes increasingly difficult to mount a meaningful defense. Memories fade. Witnesses may die. And while it’s not completely clear at this point just how the linkage between investigative genealogy and John Doe warrants will play out, the bottom line is this: every time there’s an unsolved case with DNA at the scene of a crime—any kind of crime—and you foresee a statute of limitations problem, you have an inducement to get a John Doe warrant.”
Roth, however, feels that reasonable safeguards could be put in place to ensure that John Doe warrants don’t threaten basic civil rights.
“If a state allows John Doe warrants and indictments, maybe there should also be a statutory requirement to appoint a counsel for the anonymous person being sought,” says Roth. “That way, an appointed attorney could at least start planning a meaningful defense.”
James Dempsey, the executive director at the Berkeley Center for Law and Technology, observes that John Doe warrants aren’t unprecedented. Their application to genetic data may be new, but they’ve been used for some time to specify unknown suspects linked to Internet accounts or phone numbers associated with criminal activity.
“Or say a robber has been caught on a surveillance camera at a convenience store, and a John Doe warrant for the unknown perpetrator is issued,” says Dempsey. “Everything considered, I’m not terribly troubled about [the NorCal Rapist and the Golden State Killer]. You had law enforcement accessing data from an open source site, uploading data and getting matches like anybody else. The only difference was they were using the matches for investigatory purposes.”
When it comes to new forensic tools, open source genealogical databases are just one of many emerging
technologies.
Berkeley Law professor Jon Simon believes that John Doe warrants serve a legitimate end, in that they narrow the potential range of government intrusion on the general citizenry. By honing in on a prime suspect, even one known only by DNA, they relieve the burden of potential culpability for everybody else.
“So in that respect, they’re a positive thing,” says Simon. “Also, the fact is that relatively few searches are done through warrants these days. There are now so many exceptions to warrant requirements [due to court rulings] that law enforcement can usually get what they want without a warrant.”
Simon also downplays privacy concerns in light of the seriousness of the offense.
“Rape is a heinous crime, it’s horribly under-enforced, so I think most Americans would approve of any reasonable technique or policy that expands deterrence,” he says, noting that a primary purpose of criminal law is to discourage people from committing crimes. And in the case of rape, says Simon, law enforcement has largely failed in that respect. Victims often are unwilling to come forward because of trauma or concerns that they won’t be believed.
Until recently, even if a rape was reported, a DNA match could only be made after obtaining a cheek swab or tissue sample from the suspect. Genetic databases, including arrest databases, are changing that.
“My worry isn’t so much over open source databases—it’s over criminal or arrest databases,” says Simon. “They may be racially skewed, and there have been cases of law enforcement using them to identify possible crime suspects among relatives of people who have been arrested [and their DNA archived] for one thing or another. That’s a very real concern.”
“Where are we if government demands that Facebook run a photo of an unidentified suspect through their database?” asks Dempsey.
When it comes to new forensic tools, open source genealogical databases are just one of many emerging technologies. As an example, Roth points to one company, Parabon NanoLabs, that is now providing “phenotyping” services to law enforcement agencies.
“They’ll help upload mystery DNA to open source sites and trace genealogical trees to narrow a search,” says Roth. “But they can also take a genetic profile and create an actual portrait of the subject—blue eyes, blonde hair, pronounced jaw, whatever. It’s pretty mindblowing. So far, Parabon has contributed to 11 arrests.”
Investigators have also been able to glean data from Fitbits to help identify murder and sexual assault suspects. And, Dempsey points out, facial recognition technology is being perfected, auguring a future where public anonymity is just a memory.
“The police, of course, can already run a photo of a suspect captured on a surveillance camera through state DMV databases,” says Dempsey, “but my concerns would grow if you start seeing increased pressure from law enforcement on commercial entities. The government can already force, say, Facebook to disclose everything in a known suspect’s account. But where are we if government demands that Facebook run a photo of an unidentified suspect through their database? Or that it makes an equivalent demand from private genealogical services like 23andMe?”
They could try that at any time, Dempsey says. The only question is how far they can get.
