In principle, DNA analysis has made it possible to establish to a very high degree of probability the human source of even a minute quantity of biological matter – most notably blood, semen or saliva. The science is complex, and the degree of certainty not absolute, especially when it is necessary to differentiate between twins or siblings. But it has begun to revolutionise the process not only of detection by police but of proof in court. What is particularly welcome is that it is as potent in eliminating those who are wrongly suspected or accused as it is in tracking down the guilty.

The availability of DNA analysis has already had a marked effect on mendacious defences advanced in rape cases – the not very beneficial effect of shifting the fib from identity (‘It wasn’t me’) to consent. But it illustrates the potency of this means of identification, stumbled on exactly 20 years ago at Leicester University by Alec Jeffreys.

Although the science is itself still being debated – for example, in relation to the number of features that need to be shared by the sample and the suspect in order to reduce the risk of error to an acceptably minute level, and in relation to the true margin of error – what I am concerned with here are the human rights and civil liberties implications of the procedures. My argument is that the case is growing for a national database holding the DNA profile of everyone living in or entering the country.

The present system, sanctioned by legislation, is that the police may take and keep a DNA sample from everyone they arrest, whether or not the person is charged or convicted. This has the unfortunate effect of putting the innocent on a par with the guilty. It draws a not very logical line between innocent people who have and have not passed through the hands of the police. But it does not follow that the law should be moved back to what it once was, so as to require the police to destroy their DNA records of everyone not eventually convicted. What follows no less logically is that the taking and retention of an individual’s DNA profile should not depend at all on whether he or she happens to have come into the hands of the police.

A routine sample is already kept of the blood of newly born children. Hospitals need it in order to be able rapidly to treat a number of serious conditions, and – although it will readily yield a DNA profile – the sample is held only for medical purposes.

To obtain a DNA record, nothing more than a speck of saliva is needed: the taking of a sample, in other words, does not amount in any significant sense to an invasion of bodily integrity. What matters far more is the need for coercion. In relation to people under arrest, the power of coercion is ready-made. But for the rest of us, a good case has to be made for compulsory submission even to such a non-invasive process.

When the question of the police retention of samples was before the appellate committee of the House of Lords last July, evidence was given that, of s0me 130,000 retained DNA profiles of people who had not been convicted, around 6000 had been subsequently linked to samples found at scenes of crimes, which included 86 murders or attempted murders and 94 rapes. This is not, of course, to say that the match solved the crime: there could have been many innocent explanations. But in each case it was a start. We now know that the burglary detection rate almost quadruples when DNA is recovered from the scene of the crime.

The House of Lords decided that the retention of DNA samples of people who had not been convicted was probably not a violation, even in principle, of their right to respect for private life under Article 8(1) of the ECHR, but that if it was, it was proportionate and amply justified under Article 8(2) and did not constitute discrimination contrary to Article 14. The situation the courts were and still are facing under the present legislation, however, is an illogical halfway house between the retention only of the profiles of those convicted – a manifestly inadequate measure – and the sampling of the whole population, a measure which Parliament has so far not addressed.

There are without doubt things which need tidying up first. The present police practice, sanctioned by law, is to retain the sample as well as the digitised DNA profile obtained from it. It should be enough to keep the profile on record. If it appears to match a scene-of-crime sample, the suspect will in any event have to give a fresh sample for comparison with the evidential sample. But there exists a good possibility that sooner or later more will be able to be deduced than is now possible from individuals’ genetic material. If it one day becomes possible to anticipate and treat disease by more sophisticated DNA profiling, a universal national register will need to hold samples, not just profiles. Regard needs therefore to be had to the possibility of segregating a universal database from the police database.

It needs to be a strong premise of the discussion that the DNA data on the police national computer is to be used solely – as it now is by law – for the purposes of preventing, detecting, investigating and prosecuting crime. Cross-referral for medical purposes raises other issues – for example, discriminatory insurance practices – which do not have to cloud the present question. There is a parallel case to be made for a separate national register maintained for benign purposes, such as identifying disaster victims without the distressing procedure of taking samples from their close relatives, or tracing lost or abducted children, and perhaps one day for making medical prognoses. If this were to be the sole authorised national database – and it is much harder to find civil liberties objections to it than to a police database – provision would have to be made for a restricted, and perhaps judicially authorised, linkage to the police national computer where a need could be proved.

What is the rationale of objection to a comprehensive national DNA database? Although, in a cogent partial dissent in the House of Lords case, Baroness Hale argued that the collection of data about something as intimate as a person’s genetic make-up must be an invasion of their private life requiring justification under Article 8(2), she agreed with the rest of the appellate committee that a complete justification by demonstrated proportionality was made out. That is my view too; but – as I think the whole of the House recognised – the justification would have been plainer and more logical still if the logging of people’s DNA profiles did not depend on the fortuity of their having passed through the hands of the police.

It can be, in fact, something rather worse than a fortuity. We know that there is an ethnic imbalance in arrests for certain types of offence, as well as in the use of stop and search powers. This is a serious issue which has to be separately addressed; but it has the unacceptable consequence that members of some ethnic minorities face a disproportionately high chance of getting on to the police DNA database without being convicted of anything. A universal and uniform database will at least resolve this problem.

There remains the concern about possible abuse, that the police might in future use the data not merely for detection but for personality profiling – especially since one of the purposes already sanctioned by law is crime prevention. I think this concern is real. A number of states – and there are indications that England and Wales may join them – have begun to allow the indefinite detention of sexual offenders on the basis of predicted behaviour. Apart from possession of the Y chromosome, which appears to be connected to aggression but which is a characteristic of all males, there is no hard evidence that heredity is a predictor of criminality. But the endeavour to establish genetically determined propensities is unlikely to stop, and we need to decide sooner rather than later whether we are prepared to let a police database be used for this purpose. If we are not, the surest way of preventing it is to store the digitised profile but to destroy the sample. We ought also perhaps to have something larger in mind: that the perennial risk of future abuse is a sound argument against needed present reform only if it is a risk which cannot be adequately guarded against.

The need for independent corroborative evidence does not diminish but grows with an increased use of DNA profiling. Each of us must have innocently left our DNA – perhaps a hair or a fingerprint – in places which will one day be the scene of a crime. Suspicion – proof even more so – has to be based on more than such coincidences. But where at present the only identifiable DNA will belong to people who have been arrested, with the associated risk that only the usual suspects will fall under suspicion, a universal database will ensure that the process of elimination starts from the full range of potential suspects.

Even then, no one can rule out the risk of a corrupt police officer planting DNA evidence at the scene of a crime – a hair from the suspect’s comb would suffice. But this is not a new problem, and it is not confined to genetic material: dust from a safeblowing will do just as well. Even fingerprints can be planted. And the risk of DNA planting is already there in the present system. You deal with such risks by discipline and training; not, unless there is genuinely no alternative, by abandoning the system.

What is more troubling is that everyone whose DNA is identified at the scene of a crime comes under a degree of pressure to establish their innocence. This is already true of both convicted and unconvicted people on the existing police database. But if it is a price society is ready to see paid, it should arguably be paid by all its members equally. This is the big issue that Parliament needs to debate and decide. The principal task of the courts will continue to be to ensure that, whatever the range of admissible evidence, coincidence is not confused with proof.

I readily accept that a national police DNA database will not be the end of the road. Not only will the growing range of information obtainable from DNA profiles put pressure on Parliament in future years to enlarge the uses to which the database may be put; there will be calls for separate and parallel databases dedicated to other purposes. Few if any will be able to achieve legislative backing, but a number may start to operate by a form of take-it-or-leave-it coercion: in other words, the price of securing a benefit – insurance, for example – may be that you provide a DNA sample and consent to any analytical use the holder wishes to make of it. To the extent that the adoption of a universal statutory police register of DNA profiles may help to make such moves seem acceptable, I accept that it is undesirable. But the answer for a modern society has to be tight controls, of the sort that the data protection legislation is already accustoming us to, on the misuse of information.

There is, in other words, no gain without risk; but in a society disturbed not only by serious crime but by the possibility of people being mistakenly acquitted or convicted of it, the potential gain represented by a comprehensive national DNA register is considerable; and the risks, so long as they are confronted, are controllable. I make no case for or against the introduction of compulsory identity cards; but a society which feels able, as ours does, to give serious consideration to such a step ought not to turn its face away from the case for a universal DNA register as part of a modern criminal justice system.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN

Please include name, address, and a telephone number.


Vol. 27 No. 3 · 3 February 2005

Stephen Sedley underestimates the shift in evidentiary standards that may follow the creation of a national DNA database (LRB, 20 January). DNA evidence possesses enormous power as a corroborative instrument, but is less reliable when used as the sole basis to identify suspects. The larger and more comprehensive the DNA database, the less reliable the evidence, as the number of accidental matches increases correspondingly. There will be a greater likelihood of miscarriages of justice, especially since juries will treat DNA evidence with great reverence. There is a significant risk that, with a national database in place, serious, expensive and difficult detective work will be avoided in favour of merely rounding up database matches and grilling suspects, with a consequently disastrous reduction in the standards of proof.

Thomas Davies
Guildford, Surrey

Vol. 27 No. 4 · 17 February 2005

If crooks want to avoid leaving fingerprints, all they have to do is wear gloves; it’s much harder to avoid leaving DNA. A better tactic would be to spoil the crime scene with DNA from someone else. It is easy to obtain other people’s DNA from the contents of their dustbin, and a crook could choose either to frame just one person, or to leave dozens of fake samples. If everyone’s DNA were to be on file as Stephen Sedley proposes (LRB, 20 January), it would lead to a considerable increase in wrongful convictions. Since fingerprints can be faked too, a fingerprint database does not solve the problem. All of which makes it important for identity cards, if they are to be introduced, to contain iris or retinal scans rather than DNA records or fingerprints.

Adrian Bowyer
University of Bath

Vol. 27 No. 5 · 3 March 2005

Thomas Davis (Letters, 3 February) and Adrian Bowyer (Letters, 17 February) both point out something I went to some trouble to point out in my article: that being able to identify far more of the DNA found at the scene of a crime will by itself simply enlarge the initial range of suspects, all but one or two – perhaps even all – of whom will be innocent. They regard this in negative terms, as merely sweeping more people into a net of suspicion and possibly unjust conviction. My point, however, was that insofar as this is already a risk, it affects a tranche of people unfairly marked out by having previously come to the attention of the police. As I argued, both they and the wider cohort which a full DNA register will create are entitled to rigorous protection from unfair incrimination: principally by recognising that DNA evidence by itself cannot convict anyone.

A major worry here is that juries can be seduced by circumstantial evidence which sometimes enables lawyers to put two and two together and make five. Stefan Kiszko was convicted of the murder of a child not only on his own apparent confession but on a seemingly cogent piece of corroborative evidence. The police had found in his car a piece of card on which car registration numbers were scribbled, one of them belonging to a vehicle which was proved to have been driven past the scene of the murder on the afternoon it happened. It looked damning. It was only when it became clear on quite separate grounds that Kiszko could not have committed the murder that everyone realised that what he had said all along was true: he had had a spat with the car’s previous owner at a petrol station and had scribbled its number down, as he was always doing, with a half-formed intention of reporting the driver. The car had subsequently been sold to someone else who, by pure chance, had driven it over the moors on the day Lesley Molseed was murdered.

The risk lies not in using DNA to establish who might have been present: such data would have made it impossible even to suspect Kiszko. It lies in the probabilistic world of the law, which dulls the edge of chance. Suspicion needs to be shored up by independent corroboration, but in the assessment of supposedly corroborative evidence (indeed, in the assessment of all evidence) the role of chance – the stock in trade of film and literature – deserves better recognition than it gets from the courts.

Stephen Sedley
London WC1

send letters to

The Editor
London Review of Books
28 Little Russell Street
London, WC1A 2HN

Please include name, address and a telephone number

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences