Forum:National DNA Database

< Index | General discussion

The recent events involving Goonie and Thanatos have me thinking. Is it about time the UK / USA created their own national DNA database, which everyone has to submit a record to? The majority of the opposition to the idea seems to be based on misinformation (e.g. using it for psychological profiling), so what if it was implemented using the following rules:

1. Only prints (standard DNA multiple restriction enzyme prints) are to be kept. All samples of DNA are destroyed after the print is created
2. The database is to be used only for criminal investigations, and not for civil or company /employment law
3. When a match is found, the information is used only as a lead and NOT as evidence in court (e.g. someone is raped, the DNA is matched to a person who lives near the victim, the police go and question them about their whereabouts and only use subsequent evidence including an identity parade). The actual match should be inadmissible in court.

What are the thoughts of the skeptic community? Crundy^{Talk nerdy to me} 09:45, 16 September 2010 (UTC)

I'd be bothered by the "everyone has to submit a record" part. This seems like a really huge undertaking actually, and one which raises issues we haven't really hashed out yet. One really obvious issue is: Is a part of the human genome subject to intellectual property rights, and who owns it? The obvious answers to me would either be "no" and "no one" or "yes" and "the person who has that particular genome (maybe twins split it)". Current patent law, however, suggests the legal answer is "yes" and "whoever sequences it first".

Maybe that seems like a tangential question, but you have to wonder how much information you should be required to give out about yourself before any crime is actually committed, and whether you in some sense have the right to "own" such information and control how it's used (like you own your thoughts or your personal diary or naked pictures of yourself). Even if the tissue/DNA samples are supposed to be destroyed, you are still being legally required to provide a very personal set of information, for no reason other than because you are a U.S. citizen (or legal resident). This seems very difficult to reconcile with U.S. privacy rights and the 4th amendment (which I think would make this system completely legally impossible, as the amendment is usually interpreted). This is not to mention the logistics of managing the DNA collection and the database itself, while maintaining adequate security and privacy protections. That's not necessarily impossible, but I find the prospect very daunting.

A person who voluntarily submits such information, or is convicted of (or is being investigated for) a crime seems like a different situation entirely from just sequencing kids' genomes and putting them into a national database (although people do save fingerprints from their children for identification).

Another problem involves these "matches". We know that they can give false positives (either because two people just happen to have similar "prints" or because they are closely related). We also know that police can sometimes put too much stock in certain information even if it can't be used in court. Currently DNA evidence is only used to bolster a case where there was other evidence pointing to a particular perpetrator. But if this system was in place, and a "search" was done for matching DNA, and a match was found for an address kinda-nearby, that might not be a stunning bit of evidence. In fact, it might happen a ton where the population is dense, if false positives are common enough. But it might lead to an investigators rationalizing an excessive focus on this one tool, if they have no better leads, a lot of pressure to solve a case, and a good amount of confirmation bias.

It's hard to judge the effectiveness of a system that doesn't exist yet, but I'm not convinced that this would be much more effective than the current system of "compare DNA evidence only to a database of convicts and people we already suspect". --Quantheory (talk) 10:42, 16 September 2010 (UTC)

Disclaimer: I have no idea how UK law would treat this. --Quantheory (talk) 10:43, 16 September 2010 (UTC)

For your inital points:

1. The question of intellectaul property rights doesn't seem to apply. It's the same as keeping a fingerprint record, in that you can tell nothing at all about an individual from a fingerprint, apart from matching it to another fingerprint. If I rewrote the whole post but substituted "DNA" with "Fingerprint", would your reaction be the same? Also, talking about sequencing genomes is nothing to do with this. All we're talking about is taking a print of individual's DNA which provides absolutely no information about the individual whatsoever (have a look at an example (but "arty") print here).
2. You are right about matches being occasionally inaccurate, which is why the match itself should only be used as a lead and not as evidence. This is still more accurate than relying on descriptions from victims which can, and indeed do lead to false arrests (I remember a case where a guy was held in prison awaiting trial for a year, he became a suspect simply because he walked his dog in the area of the murder)

Crundy^{Talk nerdy to me} 10:58, 16 September 2010 (UTC)

P.S. regarding pressure on police to solve a case: In the UK, since the CPS took over the role of deciding charges and whether to go ahead or not (as well as the actual prosecution), the number of large miscarriages of justice caused by police forcing confessions, using circumstantial evidence etc has plummeted. Either they get good evidence and the case goes ahead, or they don't and it doesn't. Crundy^{Talk nerdy to me} 11:01, 16 September 2010 (UTC)

"If I rewrote the whole post but substituted "DNA" with "Fingerprint", would your reaction be the same?" Maybe not on an emotional level, but actually my suspicions regarding US privacy law would be the same. Currently biometric data can be gathered from suspects and felons, or as a routine requirement in order to obtain certain licenses, clearance, or privileges. I'm certain that it can not be gathered across-the-board from everyone, for any reason, and I'm not sure that I would approve of doing that even with fingerprints. For that matter, as far as I know it's acceptable for an innocent person to refuse to allow him/herself to be photographed (at least on religious grounds), if one is willing to give up certain privileges (like a driver's license). It would take vast changes in civil rights law in order to actually mandate entry into a database like this, or even just to tack it on to an existing system.

Fingerprinting is also far more transparent than DNA, because I know what information I'm submitting when I give a fingerprint, whereas if I send in tissue sample containing DNA, I'm sending in a great deal of information and trusting that they are only going to extract and store a tiny amount of that.

Regarding the lead/evidence thing, I'm actually arguing for the opposite conclusion from you, that DNA should be used as evidence and not as a lead. I'm not so concerned about inaccuracy of the test, as about DNA planting/fraud (by criminals themselves, or less likely by corrupt law enforcement) and especially about the wp:prosecutor's fallacy. I think DNA can wrap up a case that was already decent. It might even make sense to use it as a lead when comparing to a database of felons (who are more likely to commit violent crimes). But I'm not sure that sieving through a universal database for a match is worth the costs and risks involved in creating such a database in the first place. And using it as a lead instead of as admissible evidence might remove one of the most useful pieces of physical evidence you have in some cases. --Quantheory (talk) 22:34, 16 September 2010 (UTC)