Episode 13, PII: From Pap Smear to Prison: DNA Collection and the Fourth Amendment

Professor Hank Greely, Part II

Catherine: Hello, and welcome to part two of our conversation with Professor Hank Greely! If you haven’t already, be sure to go back and listen to part one, where we discuss the Fourth Amendment and the case of the BTK Killer. In this episode of Civil Libertrees, we’ll pick back up with Professor Greely explaining how family searching, as in the case of the BTK Killer, has wide-ranging implications for your privacy. 

~Music fades out~

Professor Greely: So there are a lot of cases like that. Where you get DNA from relatives, and you can use it, you collect DNA surreptitiously from suspects. That's one class of cases. Then you can try to take crime scene DNA and connect it to relatives. That's another class of cases. And the culmination of this is genetic genealogy where there are a number of companies, three big ones, Ancestry DNA, 23andMe, and Family Tree DNA, which have been selling people information about their DNA, mainly people who are looking for relatives. And so somebody had the bright idea of trying to solve a cold case by taking  DNA, crime scene material, sending it to one of these companies, and then sending that information to someplace else and searching to see if  there were any relatives.  

And I think the first case started with a third cousin or a third cousin once removed, not close relative to if there's a parent or child or a sibling relationship. And there’s now been probably, maybe, a hundred or more cold cases around the country that have been solved this way. Now there are a lot of cold cases; a hundred isn't a big number. But they're very dramatic. 

The Golden State Killer, the GSK. It was one from California. That was one of the very earliest and achieved a lot of notoriety. There are companies that will help do this. The problem that has stalled some of the genetic genealogy is you still need a big database. Two of the three companies who've said, “No, we're not going to make our databases available for this”: Ancestry and 23andMe.   

And then there's this thing called GEDMatch, which allows you to take the information you get from one of the three companies and put it in a broad database. Right now, if you've done Ancestry, and you've got a relative on 23andMe, you won't see them because their databases are combined. But if they send theirs to GEDMatch, and you send yours to GEDMatch, it'll work. 

GEDMatch policy originally was anybody could look at anything. They actually weren't a for-profit company. They were sort of a hobby. Then when this broke, they said, okay, we'll let people opt out. And if you've opted out, we won't make your DNA available for forensic uses. And then they changed again and said, no. Only make it available if you've opted in. And I think they've got several million profiles, and I believe the last I saw, majority had opted in, but not all of them. So the limiting factor for genetic genealogy is the database.  

Last point. What are these databases, and where else can you find DNA information? Well, right now, if you've been convicted of a felony in the U.S. or you've been arrested for a felony, some immigration situations will take DNA save it for various purposes. It's authorized by federal law. 

A few counties and other smaller law enforcement areas have collected their own DNA. Orange County, California has a huge county database that can't be in the federal one because it doesn't meet the federal standards. 

Catherine: The federal standards in that, like the data wasn't collected?

Professor Greely: The federal standards are it has to be authorized by a statute that authorizes collection of DNA from people who are convicted of a felony or charged with a felony.  

If you're arrested for a misdemeanor in Orange County, and it's not a very serious one, for over a decade or so, Orange County has been saying, “Okay. We'll dismiss the charges if you voluntarily give us a DNA sample.” And they have over a hundred thousand DNA samples that were obtained this way. 

Andrea Roth, a law professor at Berkeley, has written really shocking stuff about that. I think that's the worst of the current methods. But—this is not a big deal yet. But I think soon, almost all of us will have DNA information that's in a database. Some of us already do from taking part in research. The federal government has something called dB gap, the database of genotypes and phenotypes that has, I think, 2 million DNA samples in it for research purposes only. 

I'm a member of the Kaiser health care system. I get my health care through Kaiser right now. They don't have genetic information that they've data-based on me. But at some point it's going to be in our health records because our DNA is going to be important for health, or are we going to believe it's important for our health, but at that point, you could go, maybe even get a court order a warrant and say, “Pretty sure this crime was committed in Northern California. We have good reason to believe it was committed by somebody who lives in Northern California. 50% of the people in Northern California get their health care through Kaiser. And 80% of those have some DNA evidence in the Kaiser databases. Let us search the Kaiser database.”  

So as DNA becomes more useful and more data-based, all those databases are one court order, or maybe just the voluntary agreement of whoever runs the database, from being used for forensic purposes. I think that's troubling.  

One of the most troubling things about CODIS—the one that collects people who have been convicted of a felony or have been charged with a felony—is the ethnic disproportion.  African-Americans are very heavily overrepresented in that database because they get arrested and convicted of felonies at a higher rate. The reasons for that are complicated and controversial. But it's a fact like something like 40% of the felony convicts in America are African-American, and they're only about 13% of the total population. Hispanic Americans are overrepresented, Native Americans are overrepresented. The various different kinds of Asian Americans tend to be underrepresented in the databases. So there's an ethnic bias there. 

You're more likely to be caught if you're black, and if that database is being used for family connections, you're more likely to be caught if you've got a family member who's been arrested if you're black than if you're white. One sort of wild solution to that, which I'm not sure the ACLU would like—and I think which raises serious constitutional questions—you can avoid the discrimination if you put everybody in the database.  

And putting everybody in the database also makes the family searching go away. You don't need to search for the family if every adult or teenager is in the database. There are downsides to that.  The country of Kuwait proposed that, and it seemed interesting. And then I read a little bit more about it, and there was real interest there in using that to figure out whether the husbands of women who had given birth to children were the fathers of those children, which in a heavily patriarchal society like Kuwait could lead to bad things for the women and the children involved. 

And Kuwait has this large population of non-citizens. Using the DNA to determine citizenship could end up prompting discrimination. So a universal database gets rid of some kinds of discrimination, but it may enable others. One last issue along those lines, the People's Republic of China has been collecting DNA from a lot of people in China. And particularly Uyghurs from Shenyang, Xinjiang, in Western China, and Tibetans, two populations where there  has been resistance to the Chinese government. So, there are certainly implications of all this for civil liberties in the United States and around the world.  

On the other hand, it's a really useful crime fighting tool. And defense attorneys initially hated it, until they realized it actually got a lot of their clients off. Most of the time when somebody does a DNA analysis here, not a match. And that's pretty good evidence that you didn't commit the crime. 

So it has, to the extent it makes the criminal justice system more effective, that helps people both who are innocent not be prosecuted and not be convicted as well as  helping convict people who really are guilty.  

Catherine: What do you see as the way forward to address the systemic issues with collecting people's data, especially for people of color who, like you said, are disproportionately represented in the datasets?

Professor Greely: Some colors. But not at all. Everything's complicated. 

I don't know what a good way forward is because it almost certainly at this point involves legislation rather than judicial decisions. There are exceptions. There are some state courts that might be willing to cut back on some of the uses. The U.S. Supreme Court certainly has only gotten more conservative since it approved collection from people who were charged but not yet convicted.  

A better response would be legislative action. But, right now, Congress couldn't agree on what two plus two is, let alone something as politically charged as anything about crime. Some states have limited genetic genealogy—a couple of them, not very many—some family searching, things like that, but I do think it would be good for there to be broader dialogue between civil liberties groups and groups interested in minority rights and the police and the prosecutors and pro-law enforcement groups to try to figure out and bring the stakeholders together and see if there are limits that can be broadly accepted. Otherwise I worry that we're going to back into a situation, particularly through things like the medical databases, where we get a universal database without ever realizing that's what we were doing. 

Catherine: So it seems like we really need some sort of active effort, not just passivity and trying to stop things, but also an active effort to make sure that these unconsensual databases don't arrive. 

Professor Greely: I think that would be a good thing. Although again, to put it in a broader context, DNA is not the privacy issue that I think is. Most immediate and most pressing for all of us—that's the internet and our cell phones. There are a lot of privacy issues, and I am not terribly optimistic it will come to good solutions for them.  

But it's also the case that people's use of privacy changes over time. They're different from culture to culture. In Iceland, people's individual tax returns are public information, so you can look up your neighbors tax return.  I don't think that would work in the U.S., I think there'd be a revolt. And people say that younger people are less concerned about privacy than older people… I'm not sure it's really true. I remember when my kids, who are now in their thirties, discovered that their grandmother was using Facebook. They apparently did a lot of changing of what was on their Facebook page. 

But it is true, concerns about privacy vary from person to person and culture to culture. And maybe our culture is just going to end up changing and accepting that there's less privacy than there used to be. Or maybe it can be successfully fought. 

Time will tell. 

~Outro music~

Catherine: Thank you so much for tuning in to this episode of Civil Libertrees! Stay on the lookout for our next series on environmental justice, where we discuss land sovereignty and indigenous rights. See you at the next episode!

Episode 12: From Pap Smear to Prison: DNA Collection and the Fourth Amendment

Professor Hank Greely, Part I

Catherine: Hello, and welcome back to another episode of Civil Libertrees! Today, I'm here with Professor Hank Greely. Professor Greeley teaches here at Stanford Law School and by courtesy at the Stanford School of Medicine, where he specializes in examining the legal, ethical, and social ramifications of emerging bio-science technologies. Professor Greeley also directs the Law School Center for Law in the Biosciences, and he chairs the steering committee for the Stanford Center for Biomedical Ethics. Professor Greeley graduated from Stanford with a B.A. in political science in 1974. So we're so excited to have you here with us today, Professor Greeley. Thank you for speaking with us. 

Professor Greely: My pleasure. I like to talk. 

Catherine: So with the tremendous advancements in technology and data collection that we've seen in the past years, to begin, I was wondering if you could briefly outline some of the current forensic uses of DNA and how they interact with and potentially infringe on the Fourth Amendment. 

Professor Greely: Sure. This is such an interesting field. I would note first though that we do tend to focus on DNA because it's exciting, and it's sexy, and it's the code of life and all those things, but the cell phones that we carry around are probably a bigger source of risk to our privacy than anything in our genes. 

DNA though has been used in forensic purposes for just about 30 years. In those 30 years, it's come a long way. The very first use was actually in the late 1980s in the UK, but it wasn't until about the mid nineties that it became regularly used. In those days you needed a lot of blood or a lot of body fluids or body parts. It wasn't easy to get DNA from a sample. 

Since that beginning 30 years ago, our ability to find and analyze DNA has gotten infinitely better. So now the problem isn't you need a large pool of blood in order to get DNA. You can get DNA from anything. So much so that the desk that you are currently leaning your elbow on has DNA from probably scores of different students. 

And some of it will still be viable. So in a way, our increased sensitivity means we're getting mixed samples more. Which makes life more complicated. 

We're also able to get samples in situations where we couldn't before—DNA that had been out in the snow, in the sleet and the rain, or baked by the sun, or in a fire. And now those things are becoming available. That also pushes back the timeline. 

We're getting DNA from humans who lived a hundred thousand years ago. The technical side of getting and analyzing DNA has gotten much better. On the forensic side, what's important is not just being able to get the DNA and analyze it. And also, something that's gotten more complicated, making sure you're looking at one person's DNA rather than a mix of a bunch of people. 

In the United States for again, just about 30 years. The main database has been something called CODIS: the Combined Operating DNA Information System run by the FBI. That has DNA profiles, not whole genome sequencing that serve as a fingerprint of your DNA identity.  Originally, those were only collected for some very serious crimes. Some come from federal crimes, the federal government. But states can choose to participate in. 

Eventually every state has chosen to participate. But initially many of them just said rape and murder. That's it. But then it became rape and murder and assault, and then it became rape and murder and burglary or robbery, and then it became all felonies. And then it became felonies and serious misdemeanors or felonies and juvenile offenses, which are technically criminal convictions. And the biggest recent step, from about 20 years ago, the state slowly began adopting this—DNA samples compelled from people who have been arrested for a felony, but not yet convicted. So the database has gotten bigger. The CODIS database now has a bout, it's probably closing in on 20 million different individual DNA profiles. They have the pairs of numbers, and they have a coded identity, and they do gender as well. 

And so if you are a police department and you've got crime scene DNA, or for that matter, if you've got a body that you're trying to identify, you do the CODIS markers, you submit it to the FBI, and they search their database, and they say, “No matches,” or, “Yeah, there's a match here.” 

Or sometimes they'll say, “This isn't a perfect match, but this is a really awfully good match here. You might want to rerun your DNA to see if you made a mistake in your analysis.” And then that gives you a lead and the police department then tries to follow up on the lead. That's the classic forensic uses of DNA as the database gets bigger. More people get implicated. Our ability to get DNA from artifacts, from crime scenes, from dirt, from the air, has gotten bigger. 

It used to be you couldn't get DNA from hair unless you had the whole root of the hair. It was plucked out. Now we're able to get DNA from the strand of hair. So you go to the barbershop, they don't pull your hair out. For the most part, they cut it. That cut hair didn't used to be able to provide DNA. Now it does. So those have made this more broadly useful. Also means if you're worried about people stealing your DNA, it's gotten a lot harder to prevent.  

Catherine: Thank you so much. That's really interesting—I had no clue that you could collect DNA even literally from the hair. So then could you just explain a little bit how these new and advanced methods of collecting DNA and other data could potentially infringe on the Fourth Amendment?  

Professor Greely: So the question there of course, is what does the Fourth Amendment mean?  It's a pretty short version. The government can't do an unreasonable search or seizure, which usually means at a first level, they can't search you or seize your, your body, your property, your effects. If they get a warrant, it's reasonable, and they can do anything that they've convinced a judge that they have probable cause to look for evidence. 

So the Fourth Amendment forbids unreasonable searches and seizures. If it's a search and a seizure pursuant to a warrant, where a judicial officer has said, “Yes, you have probable cause to believe there's relevant evidence there,” that satisfies the Fourth Amendment. The Fourth Amendment is one of those things where the exceptions are much more important than the basic rule, because there are hundreds of different kinds of exceptions in different circumstances.  

One of them is for stuff that you've abandoned. So if you've thrown stuff out in the trash, the police can search that, and it's not an unreasonable search and seizure because you've thrown it out. If you leave things at a crime scene—you left the gun there, you left the knife there—no problem searching and seizing that and doing the analysis of it. Fingerprints—you've left some of the oils on your fingertips on something at the crime scene—that's been okay too. 

DNA is a physical thing. It's a molecule, usually in cells. You have left it at the crime scene. And normally it wouldn't be a Fourth Amendment issue because you've left it someplace. You've abandoned it in a way. And what makes it unreasonable, I think, is you don't have a choice unless you live in a moon suit. You're leaving DNA with every breath you take. 

You can't help but give off DNA. It's not like trash where you intentionally made the decision to throw it out. You can't help it. And the courts don't really focus on that. And so you get cases where people are surprised to discover that their DNA gets matched. So one set of issues is who's in the database and how big the database is. 

The other issue is where do you get crime scene DNA, or where do you get DNA about a person? And as I noted, technically we're able to get it better than we did before. But one possibility is you don't necessarily need to go to the database. If you've got a suspect, get the DNA from the suspect. If you have probable cause to arrest somebody, you can search them for DNA. 

You can get their fingerprints, you can take their mugshot once they're arrested. There are things you can do. What if you don't want to arrest him yet, you don't think you've got probable cause? So you don't want to alert them. Then there's a whole issue of surreptitious DNA, where you get DNA from something that the person has thrown out or touched. 

They didn't know they were giving you DNA, but they were. And then you can see if that matches the crime scene DNA. If it does, do you arrest them? Because now you've got probable cause you've made a connection. And you can get a DNA sample directly from them and check to see if it actually matches. Should that be constitutional? 

Should the surreptitious collection of DNA be allowed? Courts have, basically universally. Does that make it right? I don't know, but there's another twist to it: this is a case out of Kansas called the BTK case. BTK is what this killer called himself. It was short for bind, torture and kill. 

Raped and killed 11 or 12 people in the late nineties and then went quiet. No more crimes. And then in something that you'd have to think is only true in fiction, he started writing taunting letters to the police. Bad idea. Criminals listening to this—don't write taunting letters to the police. And you remind them of your crime and you give them some clues. And so the taunting letters led them to discover that the letters had been printed on a printer that was located in a church in Topeka, Kansas. How exactly did they go from just the letter to finding the printer you have? 

They had to have suspicions about where the location was, what town it was in and check a bunch of printers. And most printers now—I don't know whether it's true of laser printers, this was old enough that it would have been probably an air jet printer, inkjet printer—but like typewriters, they all have little quirks. 

The keys on typewriters aren't all exactly the same. The ‘E’ will be a little crooked in one or have a little slash through it than another, so yeah, they could do that and they had a suspect. But they didn't have any strong evidence he’d been in the right area. He was the right age. He was male. Could have done it. But they didn't want to arrest him. They had lots of crime scene DNA, but this guy wasn't in any database. So instead they knew he had a daughter who was a student at Kansas state university in Manhattan, Kansas. A young woman. They guessed accurately that while a student, she probably had a pap smear screening for cervical cancer.  

Pap smears and other pathological path specimens get saved. And so they went to the student health center, asked if they had a pap smear for this young woman. They did. They got it from the center—whether they got it under a search warrant or a subpoena, or just voluntarily, has never been clear to me. They checked the pap smear’s DNA. And yes, it was consistent with the crime scene DNA belonging to a first degree relative of hers either a parent or a child. She didn't have any kids who were old enough to have committed these crimes.

They then arrested the father. His DNA matched. And they charged him with 11 counts of murder and other things. He pleaded guilty to avoid the death penalty. Now—this is interesting. They surreptitiously collect DNA, but not from him, from his daughter. They don't ask his daughter's permission. Maybe they've got a warrant, maybe they don't, it's not entirely clear. 

It never got litigated. And they use that to catch him. Her first reaction was horror. She couldn't believe her dad had ever done things like that. And she didn't want to have been complicit in him being arrested. Although over the years, her position has changed and she's decided that her dad really was a monster. Good dad to her. And that she's kind of glad that her pap smear led to his arrest. But mixed feelings, as you can imagine… 

Catherine: On the next episode of Civil Libertrees, we’ll continue this conversation and discuss how the confidentiality of corporations who handle your genetic material may be much more compromised than many users realize, and we’ll ask the question, “Is privacy really dead?”