I wonder what in particular it is about criminal justice policy that it attracts so many profoundly stupid ideas. The most stupid that I have heard lately is a suggestion that British police should be able to take DNA samples from children who “exhibit behaviour indicating they may become criminals in later life”.
Gary Pugh, director of forensic sciences at Scotland Yard and the new DNA spokesman for the Association of Chief Police Officers (Acpo), said a debate was needed on how far Britain should go in identifying potential offenders, given that some experts believe it is possible to identify future offending traits in children as young as five.
The article rightly picks up on some of the civil liberties issues that this raises, but I think that is only the smaller part of the problem. Certainly DNA samples raise plenty of privacy implications, as it is a blueprint of what makes you you. Whilst the most common methods of DNA profiling or fingerprinting are relatively innocuous in terms of what you can actually tell about a person, essentially analysing repeating patterns in ‘junk’ DNA, emerging methods (Short Nucleotide Polymorphisms, or SNiPs) seem to have higher correlates with phenotypes (physical characteristics). Potentially, someone who has your DNA can construct a family tree of your relationships, tell your likely sexual orientation, and determine whether you’re more likely to develop cancer. Not the sort of information most people are happy to trust the police with.
The bigger problem is essentially ignored through a fairly transparent appeal to authority: the idea that criminality can be accurately predicted from childhood.
Pugh’s call for the government to consider options such as placing primary school children who have not been arrested on the database is supported by elements of criminological theory. A well-established pattern of offending involves relatively trivial offences escalating to more serious crimes. Senior Scotland Yard criminologists are understood to be confident that techniques are able to identify future offenders.
That is, not to put it too delicately, bullshit. I’m not sure who these ‘Scotland Yard criminologists’ are, but they don’t publish in the journals I read, or present at the conferences I attend.
Certainly when you look at convicted criminals, the antecedents of their criminality — how they got to where they are — seem to be pretty clear. They tend to come from broken homes. They do poorly at school. They ‘hang with the wrong crowd’. They’re sexually precocious. And for the worst of them, their antisocial behaviour seems to have first occurred pretty early in their life.
The problem is, that doesn’t work in reverse. Even if you take all of these ‘predictive’ factors, most kids (most boys, really) who fit the profile will not go on to a life of crime. And a bunch of those who don’t fit the profile will, especially when you consider things like ‘white collar’ crimes (which, it turns out, damage a lot more people and cost a lot more money than property or violent crimes).
In short, these sorts of profiles have a very high false positive rate. They wrongly identify a lot of people as criminals. They also have a non-trivial false negative rate, which is less important, but still problematic.
Most developmental criminological theories allow for some flexibility in the creation of criminals. Problem kids put into a good environment (a better school, more caring parents, and so on) can do better, and good kids who suffer certain life trauma (drug addiction, unwanted pregnancy, etc.) can be put on a path to criminality. Different theories put this in different ways, but the ideas of trajectories (directions you’ll keep going unless something major changes) and transitions (points in life which can dramatically change your trajectory) are pretty central.
I’m not sure what the current status of ‘labelling theory’ is, but I think it’s core principle is still sound within the context of developmental criminology. How can anyone think that a police officer turning up to school to take a DNA sample from a child because they have been identified as statistically likely to commit crime when they are older might be a positive life event? How similarly will police react to two young boys who have committed some minor crime, one who had a DNA profile on record because he was born with the mark of Cain, and one who is clean? How might having this DNA record on the file affect ones future job prospects, even in the absence of any other criminal record?
Essentially this system would create a class of Gattaca-style genetic ‘invalids’, all on the basis on some criminologist’s actuarial model of risk of involvement in crime. As a criminologist, I think this is a stupid idea.
Whilst I’m not opposed to debate on issues like this, and think informed debate can be healthy, there is no evidence in the public arena which suggests that this would be anything other than a fundamentally stupid idea. I think this is just another matter of cops wanting to increase their own power at the cost of everyone else in society, and hopefully it will be recognised as such.
0 comments:
Post a Comment