DNA Retention: Stupid or Scary?
8 May, 09 | by Iain Brassington
Not so long ago, Søren posted an item on this blog welcoming the ECHR’s verdict that the UK policy of storing DNA samples from all people who’d been arrested, regardless of guilt, was in contravention of Human Rights laws.
A couple of days ago, the UK government published its response. It’s either remarkably sinister or remarkably silly.
In essence, the proposal is that, if you’ve been arrested but then not charged with, or acquitted of, an offence, your DNA can be kept for six or twelve years, depending on the offence for which you were arrested. The details are available from this page of the Home Office website. And the rationale offered is… um… Well, let’s just say that there’s a few glaringly obvious problems.
I feel a rant coming on…
OK. The first thing to note is that there seems to be a few reasons offered as to why it’s necessary to keep a DNA database of innocent people. The Home Office is keen to point out that DNA evidence can help prove innocence. Other points raised relate to detection, and to public protection. Let’s look at them in turn. It won’t take long.
(1) Proving Innocence. “The database has provided a pioneering method not only for catching the guilty but also in proving innocence,” says the Home Office. Doubtless. But they seem to have forgotten the idea that innocence is the starting point when dealing with suspects. They don’t have to prove their innocence. That’s taken as read. Now, when it comes to appeal, that is reversed. (Here the presumption of innocence lies with the CPS, I suppose.) But, that being the case, it’s not clear how retaining the DNA of those never found guilty will make the blindest bit of difference, just because we’d be dealing with someone who has been found guilty. Should it turn out that his DNA doesn’t match that from the crucial bloodstain, then that’s that. I can’t see for the life of me why it should be important that the DNA of people not guilty of anything should be retained.
(2) Detection. This baffles me. Quite why keeping a record of those never found guilty of anything should help with hunting down those who are guilty of something is beyond me - unless, of course, the tacit assumption is that we’re all suspects of everything, and those who’ve been arrested just happen to be convenient gene-donors. That doesn’t seem to sit so easily with the presumption of innocence, though. Moreover, it seems to imply that what we really should be doing is taking a DNA sample from everyone, the better to be able to nail them if they do ever happen to commit a crime. That’s a bit troubling, for reasons to which I’ll return. For the moment, I’ll satisfy myself by pointing out that, tacitly, the proposals seem to be saying that the innocent are only innocent because they’ve not been found guilty… yet. Give us time, and we’ll find something for which we can find them guilty, though.
(3) Public Protection. This is the silliest of the lot. For one thing - and I’m going to assume that public protection means crime prevention - a DNA database is only of any use on the presumption that crimes will be committed for which someone needs to be prosecuted. That is to say: a DNA database would be at its most useful precisely when crime prevention has demonstrably broken down. So that particular defence is incoherent.
The other option is that the assumption is that the likelihood of offending is inversely proportional to the expected likelihood of detection. That seems to be dubious, and distasteful. It’s dubious because most crime is not “rational”: it’s a spur-of-the-moment thing. Pub brawls and burglaries just aren’t carried out on the basis of an assessment of the risk of getting caught; and, if they were, it’d be by people who thought that they had a reasonable chance of getting away with it. Meanwhile, terrorists - and everyone really does have to bleat on about terrorists, don’t they? - are unlikely to be bothered by worries about detection. The politically violent might very well welcome prosecution, since it gives them and their cause publicity; alternatively - as in the case of recent religious terrorism - they don’t care about prosecution because they don’t really plan to survive. (Granted, terrorism often brings other crime - but that takes us back to the non-terroristic points I made.)
Besides: what kind of protection is this? A world in which noone dares break the law out of fear of detection is a world in which people are unlikely to dare to do very much at all. Secure it may be, but so is a Panopticon. That noone has anything to fear unless they have anything to hide is simply wrong. The fact that you’re under scrutiny is, in itself, a reasonable cause for alarm. If it’s true that those with nothing to hide have nothing to fear, then it should be OK for me to put a CCTV camera in your bathroom. I’m assuming that you don’t do anything particularly shameful in there. If you want to say that I oughtn’t because you have a right to privacy… well, gotcha. The same applies here. Even if you’ve not committed a crime, and don’t plan to, the mere fact that there’s a database with you on it is the sort of thing that reasonable people might well legitimately fear without thereby losing their rationality.
The other aspect of keeping arrestees’ DNA is that, while some people are justly arrested even though they’re innocent - they might be reasonable suspects - others aren’t. Just last night, there was a programme on Radio 4 that featured an interview with someone who’d been at a climate change camp, and had been arrested under terrorism legislation on the basis that a document she was carrying might be of use to terrorists. In fact, it was a recipe for ice-cream. (You can listen again until the 14th May - skip ahead to 14:40 and the couple of minutes after that.) Admittedly, the ice-cream sounds disgusting… but all the same. That’d be 12 years’ of DNA retention for a non-crime under current proposals.
I’m not sure how to end this post - it’s gone on long enough. But the idea that anyone who’s been arrested in England and Wales is liable to have their genetic data stored for six or twelve years seems to me to be deeply problematic. It’s either indefensible because it’s morally repellent, or indefensible because it’s incoherent. So: which is it? On which horn of the dilemma shall we impale Jacqui Smith, Vernon Coaker, and the rest? None-too-nice or none-too-bright?
Ach: sod it. I feel a poll coming on:
DNA database policy
UPDATE, 10.v.09: Denny, at Liberal Conspiracy, makes largely the same case, and provokes this rather wonderful response from Lee Griffin.
And, just to underscore the problematic nature of the Government’s arguments.
The proposals only deal with retention, not with the initial taking and analysing. Whatever the retention decision is in the end it will still be the rule that DNA is taken from anyone who is arrested for a recordable crime, analysed and compared to the DNA profiles from unsolved crimes found in the National DNA Database. This means that most of the amazing detections of perpetrators of old murders and rapes that politicians like to talk about are still going to happen in the future, because they occur when you take DNA from someone new to the system and compare it with old crime material, not by comparing new crime material with DNA from people previously found innocent.
Søren Holm
May 8th, 2009 at 3:03 pm
I certainly agree with your comments on the presumption of innocence. We should also not forget that the Home Office is building a very large database of fingerprints (the ID card seeks to have all our fingerprints) and, along with other government agencies, a mass of other biometric and personal information. The DNA database is just part of the Panopticon technology.
Over quarter of a century ago Stephen Toulmin wrote a paper entitled ’How Medicine Saved the Life of Ethics’. I wonder whether in another twenty five years an Informatics expert might present a paper entitled ’How Informatics Ended the Life of Ethics’. Of course such a transition was part of Bentham’s plan for the Panopticon. For those within the Panopticon, as Bentham explained, it makes no difference who is observing them - be it the Governor of the prison, his wife, his maidservant, or indeed that ‘nobody’ is observing them - the Panopticon continues to function. Morality and politics are made redundant by a Panopticon technology. Notions such as the presumption of innocence become meaningless with the repeated application of the technology. The Enlightenment construction of autonomy and privacy can very quickly be dismantled.
The technology does not work the way the Home Office claims it does. It has had some successes, but if the database was to be increased to include all of the population the number of false positives would inevitably lead to false convictions. But even without false positives we cannot always believe what counts as evidence.
There is a poor understanding of probability theory among governors and the governed (bioethics and medical jurisprudence are also not getting to grips with it). I am surprised how many people think that a DNA match within a large population provides an overwhelming evidence of guilt. As Hugh Mellor explains: “Take the extreme probabilities invoked when a defendant in, say, a rape case is identified by DNA evidence. Suppose for simplicity that the probability of a false match – i.e. of DNA samples from two different people matching – is one in ten million, and that a sample from the scene of the crime matches the defendant’s DNA. It may be tempting, given the enormous odds against a false match, to think that this evidence alone proves the defendant’s guilt. But it does not. For suppose the only other evidence about the rapist is that he is an adult male in the UK, which contains well over ten million such males. Then all the DNA evidence tells us is that the rapist is likely to be one of at least two people, of whom the defendant is one. So on this evidence the epistemic probability that the defendant is the rapist, far from being over 99.99%, is less than 1⁄2, which is too low to prove his guilt even on the ‘balance of probabilities’ needed to win a civil case, let alone ‘beyond reasonable doubt’, as required for a criminal conviction.”
For sure DNA evidence must be supported by other “evidence“ (this can be very weak), but, as in the past, the police have a propensity to fitting-up those they think are guilty. In these cases DNA evidence could not be used at a later date in a appeal to prove the innocence of an individual. At best the only new DNA evidence could be the identification of another individual or a “better more complete” match. The latter raises another set of problems, i.e. what is a good DNA match? As the database gets bigger (and indeed its fidelity) so does this problem, until eventually we are in a worse situation than before, except now we have lost the presumption of innocence.
All that aside, the very idea that we should abandon the presumption of innocence on the grounds of “public safety” smacks too much of a time when the Enlightenment went rapidly wrong and we got the Committee of Public Safety in France in 1793 (longway to go before that - hope).
Okay, there are my rants - Panopticon technology and probability theory. Keep up your rants on Panopticon technology and lets keep ethics alive.
Keith Tayler
May 10th, 2009 at 5:01 pm
I have to add to the post of Keith Tayler.
Professor Allan Jamieson of the Forensic Institute in Scotland has written a paper on “So DNA is objective?”. The answer is, effectively, no.
On doing the matching tests, for example, there is no, “The marker is above 6.3″ (or whatever), it’s simply eyeballed by a technician looking at the height of a bump and who may then pass it over to another technician for verification. This is not objective.
With the tiny amounts that can be passed on from person to person to object and still detected, it’s important that the stats are clear to any jury - and that there’s corroborative evidence for the crime.
Furthermore, if the database is used to prove innocence and corroborative evidence is also required here, then the logical outcome is that everyone should not only be on the DNA database but also on a fingerprint, footprint and fibre database. That shouldn’t cost more than a few hundred million a year…
Chris Powell
May 10th, 2009 at 6:19 pm
Chris
Thanks for that link. A good essay on this is ‘THE ART IN THE SCIENCE OF DNA: A LAYPERSON’S GUIDE TO THE SUBJECTIVITY INHERENT IN FORENSIC DNA TYPING’ by Erin Murphy at http://www.law.berkeley.edu/php-programs/faculty/facultyPubsPDF.php?facID=5705&pubID=10
Like all science it all comes back to a human or humans making an observation.
Keith Tayler
May 10th, 2009 at 10:27 pm
Luke Jostins
May 12th, 2009 at 5:16 pm