Nurse and Lawyer share DNA. All of it. That leads to all kinds of cool things like looking similar and both being awesome.

It also means that if one of us committed a felony and went to jail, and the government took a sample of our DNA for their database, and then the other one committed a crime, the government could probably figure it out.

For years, governments (especially in England but also here) have been trying to figure out a way to use our massive forensic DNA databases to catch not just repeat criminals, but the family members of criminals. And now, California has successfully caught a serial killer with a body count of at least 11, known as the Grim Sleeper, using the technique. (NY Times; LA Times).

In brief, here’s how it works:

The FBI has a database of DNA markers called CODIS. The federal government and all the states contribute DNA profiles (in the form of sets of numbers identifying the number of times certain “junk” sequences repeat, good for identification but not known to code for any observable traits) to the database. The profiles come from convicted felons — most states now contribute data from all convicted felons, though it used to be more common to use only certain violent felonies. Some states (including Lawyer’s own California) now also submit profiles from people merely arrested — that is, not yet found guilty — of felonies. All these profiles are in CODIS.

When the police have an unsolved crime, they can submit DNA from the crime scene, and CODIS can tell them if there’s a match. That is, if there is DNA from a previous felon in the database that matches. That’s not super-controversial. Where it gets tricky is if they get a partial match — a profile that’s similar to the crime scene DNA, more so than might be expected by chance, but does not match. In these cases, it’s likely (but not even close to certain) that the DNA came from a relative — a parent or a sibling. And because of the distinctive pattern of DNA shared between parents and children, we can often tell if it is in fact a parent, or more likely a sibling. Police can then essentially draw a family tree for the person in the database and see if there are any relatives who might’ve committed the new crime.

That’s exactly what happened here. The Grim Sleeper’s son was in the database. They had the partial match, determined that the son in the database had a father who would’ve been in the right places at the right times and matched a description they had, and then. . . followed him around until he threw a pizza crust in the trash and DNA-tested it.

(Which, surprising to some, doesn’t seem to be illegal. If you throw it away, it’s considered “abandoned” for legal purposes, and then the cops are free to do what they want with it. Also, in an incident Lawyer found hilarious, one state has even upheld the practice of mailing a letter to a suspect with a return envelope, pretending to be a lawyer recruiting plaintiffs for a class action, so they could collect a saliva sample.)

According to the NYT story, California is one of only two states to have an explicit policy allowing this technique; the other is Colorado. But lots of other states do it to one level or another.

This case — catching a really prolific and still dangerous serial killer — is the best possible argument for this use of this technique.

But there are plenty of reasons not to do it.

Should some of us be subject to greater suspicion than others because of what our relatives have done? That seems kind of . . . unamerican.

Innocent people might be harassed and investigated at a greater rate, especially if police are overconfident because of a partial DNA match.

Resources better spent on traditional investigation techniques might be wasted following leads generated by a partial match.

The convicted felons in the database are having their privacy invaded to an extent — they didn’t cause this new crime, and now the police are drawing up their family trees, perhaps trying to get information from them, and generally using them for something we know they didn’t do.

The database is populated disproportionately with minorities — so this technique could be said to keep minorities under a greater level of “genetic surveillance.”

Lawyer thinks this technique is too much of an invasion. It isn’t right to exploit family relationships in this way. And though it’s true that catching the Grim Sleeper is a triumph for law enforcement, I’m not sure it’s worth the price. I think we should either eliminate the policies altogether, or limit their use to extreme cases like this one. (I also think that if we’re going to have such policies, they should be passed by legislatures — not written into internal policies in DA’s offices.)

I’m glad they caught him. But this turn of events is dangerous for privacy advocates. It was a dramatic win for law enforcement, and none of the potential bad things that can come of this happened in this case. That doesn’t make them any less likely to happen in future cases — but it does make the entire thing sound like a better idea when I’m not convinced that it is.