S. and Marper

The case that changed a national database

In December 2008, the Grand Chamber of the European Court of Human Rights delivered a judgment that obliged the United Kingdom to rewrite its laws on forensic DNA retention. The case — S. and Marper v. United Kingdom — is the single most consequential legal ruling in the history of civilian DNA databases, and its reasoning still shapes retention policy across Europe.

The two applicants, identified in proceedings only as S. and Michael Marper, had each been arrested in England, had each had a DNA profile taken and loaded onto the National DNA Database, and had each been either acquitted or had their proceedings discontinued. Under the law as it stood — the Police and Criminal Evidence Act 1984, as amended — their profiles, cellular samples and fingerprints remained on the database indefinitely regardless of outcome. England and Wales operated, in effect, a blanket retention policy: once the police had your DNA, they kept it.

What the court decided

The Grand Chamber held unanimously that indefinite retention of the profiles and cellular samples of unconvicted people violated Article 8 of the European Convention on Human Rights — the right to respect for private life. The reasoning was careful and worth unpacking, because it went further than a simple proportionality finding.

The court first addressed a threshold argument the UK government had pressed: that a DNA profile is merely a string of numbers derived from non-coding regions of the genome, and that storing numbers engages no meaningful privacy interest. The Grand Chamber rejected that framing. A cellular sample, it observed, contains the full genome and carries information about health, ethnicity and family relationships; even a profile, though derived from regions not used to read genes, is uniquely identifying and can be used to draw inferences about biological relatives. The information was, the court found, sufficiently personal in character to engage Article 8 regardless of whether it was ever misused.

On proportionality, the court noted that England, Wales and Northern Ireland had the broadest retention powers of any Council of Europe member state. Most comparable countries applied retention only to convicted persons, or set fixed time limits tied to the gravity of the offence. The UK had no such limits. A child arrested and never charged could appear permanently on the same register as a convicted offender. The court found no evidence that this breadth produced investigative benefits proportionate to the interference with private life, and the judgment itself ↗ catalogues the range of European approaches the UK had declined to adopt.

The ruling was not that DNA databases are impermissible — the court was explicit that maintaining a database of profiles from convicted persons serves a legitimate aim. The objection was to indefinite retention of the unconvicted: the failure to distinguish between categories of person, and the absence of any mechanism by which an individual could have their profile removed.

CROSS-REF

The legislative response

The judgment required legislative change, and it came in two steps. The Crime and Security Act 2010 introduced a tiered retention scheme — longer periods for more serious offences, shorter for minor ones — but critics, including civil liberties organisations and parliamentary critics, argued it still fell short of the Convention standard. The Protection of Freedoms Act 2012 went further. It created the regime still broadly in force: profiles of adults convicted of a recordable offence are retained indefinitely; profiles of adults arrested but not convicted are deleted, subject to exceptions for serious offences where the police may apply to the Biometrics Commissioner for extended retention. The Act also established the Biometrics Commissioner ↗, an independent office charged with overseeing retention decisions and hearing challenges, and it required the destruction of the cellular samples — the actual biological material — that had accumulated on an enormous scale under the old regime.

The practical consequence was a reduction in the size of the National DNA Database. At its peak, the database held profiles for more than five million individuals, representing a higher proportion of the population than any other country maintained. Following the 2012 Act, several hundred thousand profiles of unconvicted individuals were deleted. The database shrank; detection rates in the years that followed showed no statistically significant deterioration, a finding consistent with research suggesting that the investigative yield from unconvicted profiles is concentrated in a small subset of cases involving serious offences where specific retention provisions already applied.

What the ruling established for the wider field

S. and Marper mattered beyond the United Kingdom for two reasons. First, it established that the European Convention applies to biometric databases not merely when data is misused, but in the way data is held — the structure of retention policy itself can constitute a violation. Second, it gave courts and legislators across Council of Europe states a common framework: retention must be necessary, proportionate, and subject to independent review; the absence of any conviction is a material fact; and the state must be able to demonstrate a genuine investigative benefit, not merely assert one.

Several member states reviewed their own regimes in the years following the judgment. The ENFSI, which coordinates forensic DNA practice across European laboratories, had already been developing guidance on database governance, and the ruling sharpened the terms in which proportionality was discussed. At the international level, the debate fed into discussions at Interpol about the conditions under which profiles shared through the gateway network could be loaded onto national databases, since a profile uploaded from a country with permissive retention rules could otherwise remain on a recipient country's database far longer than the originating country's own law would allow.

The case also reinforced a point that forensic scientists had been making in technical literature for some time: the weight a profile carries in an investigation depends on who is on the database being searched. A database that includes large numbers of unconvicted individuals changes the prior probability that any match is meaningful, because the denominator is no longer restricted to people with a demonstrated history of relevant offending. This is a statistical argument, not a legal one, but it connects directly to how likelihood ratio evidence is interpreted in court.

The reasoning was careful and worth unpacking, because it went further than a simple proportionality finding.

Where the argument now sits

The 2012 framework settled the immediate legal crisis, but the underlying tension has not disappeared. As familial searching extends a database's reach to people who are not on it, and as genetic genealogy begins to draw on profiles held in non-forensic repositories, the population effectively searchable grows in ways the S. and Marper framework did not anticipate. The question the Grand Chamber posed — whether the state can demonstrate a genuine, proportionate benefit for each category of person whose data it holds — has not become easier to answer. It has become more urgent, because the categories multiply while the legal instruments remain those written for a simpler architecture.