Winning the fight against database-statism

Good news from Strasbourg:

"Two British men should not have had their DNA and fingerprints retained by police, the European Court of Human Rights has ruled.

The men's information was held by South Yorkshire Police, although neither was convicted of any offence...

The judges said keeping the information "could not be regarded as necessary in a democratic society".

Home Secretary Jacqui Smith said she was "disappointed" by the European Court of Human Rights' decision.

The database may now have to be scaled back following the unanimous judgement by 17 senior judges from across Europe.

Under present laws, the DNA profiles of everyone arrested for a recordable offence in England, Wales and Northern Ireland are kept on the database, regardless of whether they are charged or convicted.

The details of about 4.5m people are held and one in five of them does not have a current criminal record."

Though the justification seems a little vacuous - what is "necessary in a democratic society," as a concept, is open to so many lines of attack, and it is unclear why such a policy contradicts democratic forms of government - this is nonetheless a good decision. There is no justification for retaining the DNA of innocent people who have been erroneously accused of criminal activity by the state unless there is a justification for sampling the DNA of the entire population. This policy attaches no value to individual liberty, nor considers the serious difficulties of protecting such sensitive data on a widespread scale.

The functional argument made by advocates of DNA sampling beyond convicted criminals is that it increases the probability of catching repeat offenders - because those who come into contact with the police are more likely to be offenders.

"In its submission to the court - and in parliamentary answers and ministerial statements - the government cites data, suggesting that by September 2005, the national database held 181,000 DNA profiles from suspects who would have been entitled to have their records deleted, under a "no conviction, no DNA" rule.

It says that 8,251 of these individuals were subsequently linked with crime-scene stains, which involved 13,079 offences including 109 murders and 116 rapes."

The problem with such claims are that they fail to distinguish whether the detection of the DNA of a person was a determining factor in their prosecution - and, indeed, as pointed out by Genewatch, "not all DNA detections lead to prosecutions or convictions." This explains the reluctance of ex-Home Office Minister Joan Ryan to make definitive inferences from these impressive and substantial sounding figures:

"As far as we are aware, there are no definitive data available on whether persons arrested but not proceeded against are more likely to offend than the population at large."

Moreover, as pointed out by the authors of a report published by the Nuffield Council on Bioethics,

"There is very limited evidence indeed that the retention regime of England and Wales is effective in significantly improving detection rates, above and beyond that which may be achieved by retaining only those profiles taken from individuals convicted of a recordable offence (as is the case in Scotland), or by simply searching against stored profiles, but not retaining the DNA profile indefinitely. The match rates between stored subject profiles and new crime scene profiles loaded onto the NDNAD in England and Wales, which is 52 per cent, can be contrasted with that of the Scottish DNA Database, which has a higher match rate of 68 per cent. This demonstrates clearly that the more limited retention policy in Scotland does not necessarily negatively impact upon its subsequent match rates."

More analysis of today's ruling here, with the full judgement here.