The BBC has a story today reporting on a recent consultation exercise by the UK's Human Genetics Commission (I know, I know - another day, another Commissioner....). The consultation exercise concluded that the DNA profiles of innocent people should not be kept on the database.
A quick look at the HGC's website will reveal a link to comments from members of the public who have visited a travelling exhibition co-produced by the HGC, "Inside DNA - A Genomic Revolution". The 'headline' comment used as the link is that old chestnut "If you've got nothing to hide, you've got nothing to fear". If, like me, you've heard that argument advanced by (or in support of) policy-makers, I can recommend this short paper by Daniel Solove (Professor in the Law Faculty at George Washington University). My thanks to Caspar Bowden for the link. Prof. Solove succinctly argues that the "nothing to hide" position reflects either an over-simplistic understanding of privacy itself, or a misleading portrayal of the risks of having none.
Here's the HGC's own news item about the Inquiry and report, and here's a link to their Summary of Conclusions... three pages well worth the read.
Among those conclusions, there were a number on which the inquiry was unanimous, including the view that the public is under-informed about the NDNAD and its use, both specifically (for instance, when a suspect is requested to provide a sample) and generally (for instance, when DNA evidence is presented in a court case).
It does seem anomalous that there should be no formal process for informing someone of their rights relating to the taking of a DNA sample, when one thinks of the clear boundaries within which the police must proceed, for example, when arresting someone, taking evidence, or executing a search warrant.
The HGC's Chairman, Sir John Sulston, described the basic issue as being one of clarifying the purpose of the NDNAD. Is it, he wonders, an "identity database" or a "criminal database"? Again, given that UK data protection law is fundamentally based on notions of "purpose of collection" and "purpose of use", isn't it both strange and dangerous that the NDNAD should have been in operation this long with that question still apparently unaddressed?
The inquiry also returned 'majority' conclusions on a number of questions, including the following:
- That the DNA profiles of those not subsequently charged or convicted should not be retained (as is already the case in Scotland);
- That the DNA samples from which those profiles were generated should, likewise, be destroyed if the individual is not subsequently charged or convicted;
- That UK samples and profiles should not be contributed to any international foresnic DNA repository, or be the subject of automatic data-sharing with other countries.
Not unexpectedly, the Home Office representative quoted in the BBC piece expresses a slightly different view, saying that the database has 'revolutionised the way the police protect the public' and giving a number of statistics relating to the NDNAD's use by the police. These related to about 11,500 cases in which 'DNA evidence was used in police investigations' into offences ranging from rape and violent crime to domestic burglaries, but - strangely - not mentioning murder at all.
Perhaps this is just another piece of fallout from Gordon Brown's June speech to the IPPR, which so recently did so much to cast doubt on the statistical basis for quantifying the benefits of the NDNAD as a forensic tool.
![[RSS Newsfeed]](/themes/sotto/images/valid-rss.png)
![[This is a Roller site]](/themes/sotto/images/roller-logo.gif){kind=logo}
Does "kept" mean "stored temporarily during an investigation" or "retained indefinitely"?
Does "innocent" mean "not yet convicted" or "cleared of all charges"? Or perhaps "free of original sin"?
There are of course two sets of samples. Those samples collected at crime scenes, to which some additional information may be attached as part of the process of identifying the person it might be associated with. And those samples collected from suspects and witnesses, which may be matched or mismatched against the first set of samples.
Presumably the Commission was primarily concerned with the second set of samples. But the first set of samples might also raise data protection issues.
Furthermore, in some contexts, the effectiveness of the database apparently relies on retaining "innocent" samples. For example, the evidence supporting a conviction for a violent bank robbery might involve many DNA samples, some of which are from the bank clerks who were beaten up. So if the "innocent" DNA is destroyed, can the courts no longer differentiate the "guilty" DNA?
It seems to me that invasion of privacy is so deeply entrenched in this kind of scheme, that it will be impossible to regulate it properly without killing it completely. Some people might think that would be a good thing, but is that every going to happen?
Posted by Richard Veryard on July 30, 2008 at 02:46 PM GMT+00:00 #
Thanks Richard - I agree... except for some confusion about your 'innocent DNA' point. I *think* what you're getting at might very neatly encapsulate the whole argument about what the NDNAD is for in general, and what individual samples are for in particular.
The phrase "to eliminate you from our enquiries" is one I associate with the collection of DNA samples from those caught up in an investigation but not necessarily otherwise involved in the crime in question. If that's the true 'purpose of collection', then surely there can't be any argument for retention once the person has been 'eliminated from the enquiry'...
Posted by Robin Wilton on July 31, 2008 at 12:25 PM GMT+00:00 #
Ah, but if I were the defence lawyer representing someone who was being prosecuted on the basis of DNA evidence, the fact that much of the relevant evidence had been "eliminated" might make it more difficult for me to argue for my client's innocence. For example, there might be another person with very similar DNA to my client, and an only slightly better alibi, whom the police have chosen to "eliminate" from the enquiry. Shouldn't I be permitted to put this evidence to the jury or the appeal court?
The deeper problem here is that databases, police investigation and common law each operates according to a significantly different logical basis, which we might call "epistemology". Boole, Bayes or Bentham?
Posted by Richard Veryard on July 31, 2008 at 01:33 PM GMT+00:00 #
I think, as you said in your initial comment, this may point to a difference in the appropriate treatment of 'evidence samples' collected from the scene of crime and 'enquiry samples' collected from people who fall into the ambit of the enquiry.
The HGC's recommendations included some to the effect that members of the public (who might end up being required to give samples, or on the jury in a trial dependent on DNA-based evidence) should be better informed in the first place, should be clearly told what rights they have in the event they are asked for a sample, and should be further informed by independent specialists on the niceties of DNA-based evidence before expert witnesses are called to argue for one side or the other...
You're right... the epistemology varies by 'community'; what I found heartening about the HGC report was that it called into question what seems to have been an unchallenged primacy of the pollice epistemology to date.
Posted by Robin Wilton on July 31, 2008 at 02:11 PM GMT+00:00 #