Justice 2 Committee, 21 Mar 2006

Meeting date: Tuesday, March 21, 2006

Official Report 278KB pdf

---

Police, Public Order and Criminal Justice (Scotland) Bill

Agenda item 2 is the Police, Public Order and Criminal Justice (Scotland) Bill. The committee will consider oral and written evidence on Paul Martin's proposed amendment to the bill, prior to it being moved and considered at next week's meeting. The six written submissions that have been received have been circulated to members, together with Scottish Parliament information centre briefings.

I welcome the witnesses for the first panel, who are Dr Helen Wallace, from GeneWatch UK and John McLean, director of the Scottish Criminal Record Office. The first question points back to the written evidence that GeneWatch UK kindly submitted. It would appear that GeneWatch UK is not in favour of the aim that underlies Paul Martin's proposed amendment. Can you explain your reasons for that?

Certainly. GeneWatch UK supports the use of DNA as evidence in specific investigations and the use of DNA databases, so our concern relates to the retention, particularly the permanent retention, of both DNA samples and DNA profiles.

We have three key areas of concern, the first of which is whether additional information about an individual that is not related to identification, such as relatedness or other genetic information, might be revealed. Secondly, there is the issue of whether keeping permanent records builds up, in effect, a permanent list of suspects that might also be misused in the future. Thirdly, there is the issue of discrimination and whether the database becomes, as it has done in England and Wales, much more inclusive of certain ethnic minorities than of others.

Obviously, you have a United Kingdom perspective on what has happened down south, where there have been changes in the process. We are looking carefully at the lessons that can be learned. What lessons would you say have been learned so far?

We think that the key lessons are in relation to achieving a balance between preventing and detecting crime, and protecting civil liberties. There has been little evidence of permanent retention making a significant contribution to the detection or reduction of crime. On the other hand, there is evidence of increasing public concern and of requests to be removed from the database, which could be counterproductive in terms of trust in the use of the database. Finally, there is evidence in relation to whether the safeguard of limiting DNA use to the purpose of the prevention and detection of crime is sufficient. We would argue, as we have done in our written evidence, that it is not sufficient because it allows, for example, certain types of genetic research to take place that undermine individuals' rights to refuse to consent to involvement in controversial research topics.

Are there any specific points that you would make about Paul Martin's amendment?

I refer to the points that we made in our written evidence. There are specific examples in England and Wales of research taking place, in spite of the existing safeguard, that is trying to predict ethnicity from DNA profiles and samples. There is evidence from the Home Office, which we have analysed, that suggests that crime detection has in fact gone down since the permanent retention of the DNA of people who are arrested but not charged. There has not been a significant increase in the detection of crime, and there is clear potential for building up a list of suspects, which could be misused in the future.

Thank you.

I ask Mr McLean to describe, for background evidence, the SCRO's current role in relation to fingerprints and DNA samples.

The SCRO retains the national fingerprint collection for Scotland, which will shortly migrate to a national fingerprint collection for the whole of the UK. That national collection—it is called IDENT1—will have the facility to retain and search nationally not only fingerprints, but palm prints, which are not currently searched automatically.

DNA is not stored at the SCRO. In fact, DNA does not come near us; it is stored at a laboratory in Dundee and I understand that records are kept at the national register down in Birmingham. The SCRO's involvement with DNA is simply to record whether a person has had DNA taken.

The SCRO's other important role in relation to both fingerprints and DNA, under the current rules, is that if someone is not convicted of an offence, their fingerprint record is destroyed; we are told that the DNA sample, too, is destroyed and we delete that record from our files.

Do you get direct evidence of that destruction from other departments?

When the individual is not proceeded against or is dealt with by means of a non-court disposal, or when a young person appears before the children's panel, which is not strictly a conviction, we are told to delete the record as the DNA has been destroyed.

So you receive formal instructions.

Yes.

Thank you.

GeneWatch UK said that there was very little evidence that the new English regulations on the retention of DNA were leading to more detection of crime. Helen Wallace said that crime detection had gone down, making it sound as if there was some correlation between keeping the DNA and the fall in the detection level. However, this is surely one of those post hoc ergo propter hoc arguments that do not actually hold water. I would like to hear more evidence for what you have said.

I did not mean to imply that keeping DNA had somehow caused a reduction in detection rates. We looked at the Home Office report that was published in January and it is indeed difficult to unpick the exact consequences of the different changes that have happened as a result of the DNA expansion programme in England and Wales. We would therefore like there to be an independent assessment that tries to do that unpicking.

We found that there has been an increase in the number of detections as a result of more DNA being taken from crime scenes, but the increasingly large number of individuals who are being included on the database seems to have had very little effect. We would expect the number of detections to rise as the number of crime scene samples that are taken goes up, but the detection rate—or the likelihood of finding the relevant individual—should go up if increasing the number of individuals on the database is important. However, that does not seem to have happened. The Home Office report says very clearly that the number of crime scene samples, rather than the increasing number of individuals, seems to be driving the success of the database.

You are implying that there is no firm evidence as yet.

I think that there is evidence from those figures. I would not say that the evidence is totally conclusive, because there are problems with trying to unpick the different things that have happened. For example, the definition of detections has changed since the database started. Detailed figures have been released only for the past two years, but not during the whole programme. A range of factors, therefore, makes it difficult to be very exact about the consequences.

Of course, comparing the situation in England and Wales with the situation in Scotland introduces additional differences. If someone is detected through intelligence-led policing and other methods, the database might have made less of a contribution. We cannot be very exact about the pros and cons of the evidence, but we can say clearly that keeping convicted offenders and more crime scene samples on the database has made a significant difference. Keeping the DNA of people who have been arrested but not charged appears not to have made any noticeable difference to detection rates.

But there has been some difference.

It has varied from year to year, so it is difficult to be exact. From the figures that have been released, it is not possible to show whether there has been any real difference. We would expect some burglars to have been identified through cold hits on the database, but it is difficult to tell whether there have been any prosecutions for violent crimes as a result of the policy.

We have to be very clear that there is a difference between taking samples from more people, entering them and checking whether those people have committed a crime in the past, and retaining those profiles to see whether they commit any crime in the future. From the data that have been provided, it is not always possible to tell the difference.

So there might be a time lag before we can prove that there has been any significant difference in detection rates. We might have to wait for five years or so.

There might indeed be a time lag, but it is clear that there has not as yet been a sudden increase in detection rates because of the policy.

Helen Wallace mentioned concerns about the Home Office figures and asked for independent research to be carried out. If the independent research proved that the Home Office figures are incorrect, but that the method is more effective than they suggest—which is a possibility—would you support the retention of all DNA samples? Is your organisation against the principle, or are you just concerned about the figures? If the figures were different, would you be happy with the measure?

Our view is that society has to make a decision and that it is difficult to balance the threats to civil liberties—we think that there are potential threats to privacy and rights—with the contribution to reducing crime. GeneWatch does not have the final answer, but we argue that we need a much more public debate that includes information and, for example, research on public attitudes to the idea of retaining samples. A range of policies could be adopted as a consequence of that debate, which might result in more profiles but fewer samples being retained, or other compromises.

Sure, but I want to ask the question again. In principle, is your organisation opposed to the retention of all DNA samples? You are happy for DNA methods to be an important part of crime detection. I am asking about a hypothetical situation. Let us say that, after the research is done and an independent report is produced for the Home Office, it is proven that, contrary to your organisation's view, the method is effective. What would your organisation's view be then?

Our view is that, in principle, we must acknowledge that retaining profiles and, particularly, samples infringes on individuals' civil rights. We accept that it can be justified in some circumstances—for example, we agree that it is justified in the case of convicted violent criminals—but the balance between the number of profiles that are retained and the number of crimes that are solved is a matter for public debate. We think that, given the existing information, retaining all the profiles of and samples from people who have been arrested but not charged or convicted is going too far, regardless of exactly what figures emerge. However, we might accept that, in some situations, some DNA of some unconvicted people in certain carefully justified circumstances should perhaps be retained. I would not like to give a final view on that.

I want to pick up on that point. You say that, in your organisation's view,

"some DNA of some unconvicted people"

could be retained on some occasions. Can you give the committee examples?

Yes. When we did our report on the DNA database in England and Wales, the preliminary conclusion—although we felt that more information and public debate were needed—was that the original policy of removing people's records from the DNA database at the same time as they were removed from the police national computer was sensible. The police national computer records are held within time limits, which vary according to the seriousness of the offence. Individuals who are unconvicted are removed from the computer within time limits, although there are exceptions in which the records are held for longer. That happens in relation to sexual offences in certain specific circumstances, with the approval of, I think, a superintendent—I forget at which level. It was agreed to hold records for somewhat longer in those circumstances. Such a policy is at least worthy of debate.

You say that it is at least worthy of debate, but could you be more specific about the sexual offences in certain circumstances that you mentioned? Could you give us an example?

I forget the exact rule. We do not think that GeneWatch should decide the rule.

I just want your view.

I believe that the rule was that records were retained in relation to sexual offence cases in which there was no doubt about the identity of the person who was involved but there had been a dispute about consent—it was along those lines. The retention was not permanent but time limited, so that the police had access to the record for a certain period.

I want to be clear about this. GeneWatch takes not a principled view, but a pragmatic view—I do not mean that to sound critical. In your organisation's view, there might be exceptions, one of which you have just touched on.

Yes. We acknowledge that there is a difficult balance to be struck between preventing and detecting crime and protecting civil liberties. I think that most people would share that view.

What is GeneWatch's view of the House of Lords ruling of 2004, which said that, in England and Wales, retention of samples does not contravene the European convention on human rights?

We disagree with that ruling. We agree with the basic principle that we must consider whether human rights are engaged and whether there is a proportional impact on them. I believe that the case in question is going to the European Court of Human Rights and is still awaiting a hearing on admissibility. The arguments that are being made in that context relate to whether the lords appreciated fully the distinction between DNA profiles and samples, which can reveal a lot more information, and whether they appreciated fully the range of people who might gain access to them in the future, for example through links with the police national computer. In Germany, a different view was taken of what is in essence the same principle. We do not see the issue as having been resolved by the House of Lords.

I am obliged.

Good afternoon, Mr McLean. What is the SCRO's view on the greater retention of prints and samples and why does it hold that view?

I should preface my remarks by saying that I am not the Association of Chief Police Officers in Scotland spokesperson on DNA; that is Chief Constable Paddy Tomkins from Lothian and Borders police. The SCRO takes the straightforward view that the more DNA samples that are held and the more DNA marks that are lifted, the greater likelihood there is of detecting offenders and eliminating from our inquiries people who are innocent of crimes. The greater retention of prints and samples helps with the detection, prevention and investigation of crime.

Does your organisation have any doubts in relation to implications in respect of the ECHR, or does that not come into the SCRO's thinking?

We are always aware of ECHR issues, but, given the House of Lords ruling, we are content that any retention of fingerprints or samples would be lawful.

I have a brief question for Mr McLean, which follows on from Mr Butler's question. The rationale behind what you just said would also stand if every citizen were to be included in a national DNA database and if we had a policy of DNA sampling every child born in Scotland. That would have the same basis as the the-bigger-the-better argument.

I prefaced my reply to Mr Butler by saying that I am not the ACPOS spokesperson on DNA. The logic of what you are saying is correct, but, as Dr Wallace said, some things are perhaps moving too far.

That is the subject of the committee's deliberations.

One of the indicators of whether we are going too far might be the use of the database and the information that you police. In an article in the New Scientist, Alec Jeffreys, who invented DNA sampling, said:

"As a geneticist I would greatly value the potential enormous power of the database for research … But it is a gross infringement of civil liberties."

Since he said that, 10 research projects have used the national DNA database. A number of parliamentary questions have been asked about the involvement of the SCRO and the Scottish Executive in the research. What is your view on the use of both the samples and the fingerprints, for which you have a responsibility, for research purposes? I noted that one of the projects was for information on ethnicity. Is that one of the indicators that you think would be going too far?

I have not considered that, but if research is done properly, is objective and is anonymised to protect individuals, it will have some value.

I have a question for Dr Wallace, which follows on from the thrust of Mr Martin's question. I understand that the strongest rationale for a change in the Scottish procedures is the Home Office figures on crime levels, detection and solving crime. In an earlier report by GeneWatch UK, you indicated that the Home Office estimates that 49 per cent of DNA matches lead to a detection, but that 58 per cent of those matches are termed "cold hits". Can you explain what a cold hit is?

The database's added value in an investigation relates to whether it introduces a new individual into that investigation. The database retains individuals' data, and if an existing individual has been identified by other means their DNA sample can be compared with a scene-of-crime sample. Cold hits are matches between a DNA profile from a crime scene and a DNA profile from an individual. By definition, a cold hit introduces a new suspect into an investigation. It is evidence that they were probably—although not with absolute certainty—present at the scene of the crime, but it does not provide any information about when they were there or other factors, so not all such matches lead to successful prosecutions. In many cases, the evidence will simply indicate that somebody had been there earlier in the day.

If my understanding is correct, the current practice is to use cold hits. If there is a crime scene DNA sample of a case that has not been solved, someone who is being sampled for a suspected offence and having their fingerprints taken will then be matched on the database against the data from the crime scene. Is that what happens at the moment in routine checks for both fingerprints and DNA? Perhaps John McLean would also like to respond with his professional view.

Some of the matches result from an individual being investigated for a specific offence; those matches are not described as cold hits, because they arise from the individual having been identified as a suspect. The added value of the database lies in the matches that result from cold hits and that can be useful in investigating a past crime, but such matches have their limitations.

If that is the case with regard to cold hits, one would expect there to be a tailing off in the number of previous crimes that could be solved or in which a detection or link could be made. That could point to the fact that, as you indicated in your written evidence, the number of crimes that were detected using DNA fell in 2004-05, even though the database grew by 1 million records.

Absolutely. Our view is that the usefulness of adding more individuals has tailed off, because the people who are being added are no longer persistent offenders.

I would like to comment on that, particularly in relation to fingerprints, although I do not think that the issue of DNA is completely unrelated. I am not familiar with the expression "cold hits" and I am not a fan of "CSI: Crime Scene Investigation", but when a crime is committed and a mark is left on the scene, a fingerprint or DNA sample may be taken. If there is a suspect at that stage, or if someone is arrested for the offence, the search will be straightforward. In other cases, a mark may be lifted from the scene of a crime for which there is no known suspect and no person in custody, and that search would be done through a database. Such a cold hit can be valuable, and numerous people might be arrested or a great deal of evidence could be gained in that way.

With regard to the point that was made about the tailing off of the value, people move through criminality during their lives. Often, people who have been active criminals—housebreakers or robbers, for example—during the ages of 16 to 30 disappear from the system and a new generation comes in. If that generation of new DNA comes into the system, that will continue to help with the prevention and detection of crime.

I understand that, but one would expect that the growth of the database to include 1 million people of all ages, types and backgrounds and in all areas would have had some impact. However, in 2004-05, the database grew by 50 per cent but there was no discernible impact on the detection of crime.

That point is debatable. Some of the figures that I have seen show that there has been a discernible impact on all areas of crime and that a significant number of crimes have been detected as a result of the techniques. It is worth noting that DNA techniques are relatively new and that officers are only starting to use them much more to populate databases.

Have you examined the figures in detail to determine how many would be cold hits and how many would not be? I wonder whether there might have been some double counting. Did you check that?

I have not done any research, personally. As I said earlier, I am not the ACPOS spokesperson on DNA.

Dr Wallace, do you agree that your analysis of the Home Office's report shows that there has been a great rise in the number of samples—as Jeremy Purvis said, 1 million samples is a considerable amount—but that there has been only a small increase in the rate of crime detection? I am sure that the committee is aware that DNA does not solve crimes on its own and that there has to be corroborating evidence, but I would like you to comment on whether that is in keeping with what was claimed beforehand by the proponents of the extension.

The figures show that the significant increase in the number of samples retained has had only a very small effect in terms of increased crime detection. The only explanation that we can give for that is that most of the claims that were made about the major benefits of permanent retention were based on the number of DNA matches, which gives no indication of the number of successful prosecutions.

For volume crime, we can work out a rough relationship between the number of matches and the number of detections from past use of the database. Unfortunately, it is impossible to do even that with regard to violent crime. We think that a misleading impression has been given that large or significant numbers of crimes have been solved as a result of the policy. The numbers appear to match not crimes solved, but DNA matches.

Mr McLean suggested that different studies would suggest a different outcome. Are you aware of any detailed studies that have been done in this area other than the Home Office one that you have put in front of us?

No. We used the Home Office report because it revealed further information that had not previously been published. An annual report for the national DNA database has been published in England and Wales only during the past couple of years. Prior to that, pieces of information were revealed through answers to parliamentary questions and so on.

I have talked to people who are doing research in this area and, as far as anyone can tell me, there is no other public information available about the value of the database. That is why the House of Commons Science and Technology Committee concluded that it would be timely to have such an investigation into the database's contribution to the prevention and detection of crime.

Maureen Macmillan said that it might be too early to draw a conclusion. As the Home Office report covers a five-year period of expansion, would we have to wait until 2010 for the next one, or is the programme an on-going one that the Home Office will monitor as we go?

The programme is on-going, but it is not clear when or if there will be a further report. That is problematic, and it is problematic that it is not entirely possible to distinguish the contribution of the different changes in the figures that have been made available. We hope that the Home Office would, at some point, take on board the recommendation of the House of Commons Science and Technology Committee, produce those figures and answer the question that you asked on the timeframe within which it would expect a noticeable effect.

In your written evidence, you say that we should

"make the National DNA Database's governing body more transparent and democratically accountable in Scotland".

What do you mean by that?

The governing body has been criticised—not only by GeneWatch, but by the Human Genetics Commission, the House of Commons Science and Technology Committee and, previously, the House of Lords Science and Technology Committee—for not being sufficiently representative and open. It is changing its composition as a result of the change in the database's operation that is being introduced. Oversight of the database is in the process of being transferred to the Home Office, but it is unclear exactly how that new system will operate.

An ethics board has been proposed to oversee research applications because, until now, there has been no such ethical oversight. However, that proposal has been under discussion for more than a year and the board is yet to be put in place, so there is clearly a lack of independence on decisions about non-routine access to and uses of the database, which include research uses and other applications.

You say that the governing body is unrepresentative. Who is represented on the body and who would you like to be represented on it but is not there?

Currently, there are representatives of various police forces and the Home Office. Until recently, the Forensic Science Service also had a representative. The new board will include two members from the Human Genetics Commission, which is an increase on only one member, which is an increase on no members a couple of years ago. That is definitely a move in the right direction, but those new systems have yet to be put in place and we have yet to ascertain whether they will provide independent oversight for decisions on the uses of the database.

Does John McLean have a different approach to collecting fingerprints and DNA samples or is it the same?

My principal responsibility is for fingerprints. My responsibility for DNA samples is to keep records on the criminal history system in Scotland of the fact that DNA has been taken and to ensure that that marker is removed when the DNA sample is destroyed in the event that someone is not convicted. The approach is the same for both.

Earlier, you suggested that I had considered various studies. That is not the case. I am aware of the Home Office report and it is perhaps a question of my interpreting it differently from Dr Wallace. The report says that 198,000 profiles were taken and that approximately 7,591 were linked to crime scenes and 11,000 offences, including 88 murders and 45 attempted murders. That gives me confidence that the database is a powerful tool for the investigation and detection of crime.

I am grateful for that clarification.

Some of the areas on which I was going to ask questions have already been covered. However, Dr Wallace, I will run the risk of asking you about the distinction between DNA samples and DNA profiles. You reflected that distinction in your evidence and the Human Genetics Commission has also picked it up. I understand that no medical or predictive genetic data would be held if the samples were disposed of but the profiles were retained. First, do you agree with that distinction? Have I got it right? Secondly, if the samples were disposed of, would you have fewer concerns about the retention of the profiles?

You have understood the distinction perfectly. It is quite important because if one can return to an individual sample, one can provide much more sensitive genetic information. That has happened in a research context. Like the Human Genetics Commission, we would be reassured if samples were destroyed. However, sensitive information could still be obtained if only profiles were retained. In particular, the comparison of profiles can reveal cases of non-paternity and can clarify matters of relatedness, whereas fingerprints, for example, do not have such sensitive information content. It is not the case that profiles are entirely free of sensitive information.

The idea that the record itself should be retained, regardless of whether it consists of a DNA profile, just a name or a name linked with a fingerprint, is still a cause for concern because it means that, essentially, there would be a permanent criminal record for everyone who had been arrested. Our view is that all samples—whether from convicted or unconvicted people—should be destroyed at the end of an investigation and that time limits should be imposed on profile retention.

If I have picked you up correctly, some of the additional safeguards that you talked about earlier would cover the retention of profiles adequately.

We think that having time limits on retention is an important safeguard, regardless of whether other safeguards are introduced. In effect, one needs to adopt both approaches. Restrictions are necessary not only on how long records are kept in different circumstances but on how sensitive information is kept and what oversight of that information there is.

I have a tiny, techie question for Mr McLean. When it comes to getting a match or assisting in the detection of crime, is there any distinction between the retention of DNA samples and the retention of DNA profiles?

I am afraid that I am unable to answer that question because it does not relate to my area of responsibility.

Fine. The answer will remain a mystery.

We might manage to think of someone to whom we can write to find out that information. That is a task for the clerks to undertake on our behalf.

People who advocate retention seem to be putting about the notion that the cold hits that have been mentioned amount to the detection of new criminals—in other words, people who have not been convicted of any crime. I would like to know what both panellists think about that. Do you agree that that is the impression that the advocates of retention are giving? Is it possible to tell whether the cold hits involve people who have never been convicted of a crime or people from whom samples have already been obtained because they have been convicted of a crime that is unrelated to the crime that is being investigated? Alternatively, do the cold hits lump together people in both those groups?

The figures lump everyone together, which makes it difficult to decide how effective the retention database is. The Home Office report cited a study in which it was claimed that a specific number of samples were followed through to find out what happened, which allowed the distinction to be unpicked. I think that I am right in saying that roughly half the cold hits introduced a new suspect, but that figure is probably relevant only to volume crimes, which comprise the majority of cases. We still have no way of knowing the extent to which cold hits introduce new suspects in cases of violent crime. We know from the report and from other material that, for a range of reasons, that process is likely to be much less useful for violent crimes. That is partly because there is usually a list of known suspects for such crimes.

I want to clarify something. When you talk about "new suspects" being introduced, do you mean people who are on the database, but who have never been convicted of any crime, or do you mean—

I am sorry. By "new suspects", I meant people who were not already under investigation for the crime in question.

My question is whether we can tell that the addition of a million people to the database has been worth the time, effort and money involved. By "new suspects", do we mean brand new suspects who had never previously been convicted of a crime and who were identified only because they were on the database or do we mean people with a previous conviction who were new suspects only in a particular case? I do not think that anyone is arguing against the retention of DNA samples of those who have previous convictions.

Again, you are right that we cannot tell. We can tell only from the overall figures that we discussed earlier that detection rates have not increased as that extra million people have been added to the database. That suggests that the chances of finding the right person have not increased despite the addition of that extra million people.

I have a slightly different viewpoint. In Scotland, the fingerprints and DNA evidence of people who are not convicted are destroyed, so we have no evidence against which details can be checked. The apparent evidence comes from the previously mentioned Home Office report, which suggests that 7,500 individuals have been linked to the crime scenes for almost 11,000 offences since 31 March 2005. Some of those crimes and offences were very serious.

How many of those individuals were on the database simply because their samples had been taken even though they had never been convicted of an offence and how many of them were people with a previous conviction?

My reading of the matter is that those individuals are people whose DNA profiles would have fallen to be removed. The report states:

"it is estimated that approximately 198,000 profiles that would previously have been removed were retained on the Database. Of these … 7,591 profiles have been matched with crime scene samples".

What is Dr Wallace's opinion?

The 198,000 figure relates to profiles that would previously have been removed, but there are a number of difficulties with it. After the first change in the law in England and Wales, the police retained the DNA profiles of people who had been charged and acquitted but not of people who had been arrested and not charged, for which we do not have figures. The figure is only an estimate because, in England and Wales prior to the change in the law, a large number of DNA samples—nobody knows how many—were retained and were not removed in accordance with the legislation that existed at the time. Therefore, it is difficult to say how the figure has arisen and where the estimate has come from.

Another problem is the figure for the number of matches with DNA at the scene of crime. We are given no information on whether such evidence has led to successful prosecutions. From our knowledge of other figures, we can estimate that such evidence has led to successful prosecutions in, for example, some burglary cases. However, it is questionable whether it has led to any successful prosecutions for violent crime.

Does Mr McLean not agree that the law of diminishing returns applies in this case? The proposal seeks to breach a basic civil liberty about people being presumed innocent by retaining their DNA profiles on a database for the benefit of the state. Adding more innocent people to the database will do little if anything to increase the crime detection rate, given that those who are involved in crime are mostly known to the police. In any community or local authority area, the police know the vast majority of those who are involved in crime. Therefore, the benefits of retaining such details are so small as to be almost negligible. Should such a fundamental civil liberty be taken away for what appears, at best, to be a negligible advantage?

I am not sure whether the law of diminishing returns applies, but anecdotes and stories appear in newspapers about how people who had not been related to a crime scene were detected because of fingerprint or DNA evidence. In each case, that is potentially one less victim of crime. If the figures in the Home Office report are accurate, they are quite compelling. As far as I am concerned, it is important if there are 88 fewer murders. I note the issue of human rights, but that is obviously a matter for the committee rather than for me.

We have discussed the figures from the English and Welsh experience. I wonder whether we should seek some other opinions on the outcomes that we can consider alongside the views that we have heard today. We thank you for your comments, but there seems to be a requirement for cross-referencing. Before we discuss the issue next week, we might have to get some information quickly about the clarity of the figures. Many people have tended to go down the route of referring to the English and Welsh experience.

Dr Wallace said that the increase in the size of the database has not led to an increase in the detection of offences. Is there an argument that the increase in the size of the database could be a groundbreaking crime prevention process? Could it be that people who have been added to the database no longer commit crimes because they are on a database?

The database could play a role in putting people off committing crimes that they would otherwise commit, but we have no information about the extent to which it does that. There is a credible argument that the fact that some repeat offenders are on the database may deter them from committing crimes. However, it is difficult to say that keeping the DNA of someone who has been arrested and never charged with an offence or who has been acquitted of an offence has deterred them from committing another offence. I find that a difficult argument, given that retaining those people's DNA has had little obvious effect on crime detection. That tends to suggest that they were not destined to commit future crimes.

But you cannot rule out the possibility that someone who has been added to the database is, as a result, more careful about their activities.

I cannot rule that out. It is a possibility, but I find it difficult to accept that that is the whole reason why the addition of a million people to the database has not had a noticeable effect on the crime detection rate.

Can I ask John McLean about that point?

Sorry, could you repeat the point?

What effect might adding people to the database have on crime prevention?

It is well known that people may be deterred from committing crime if there is a greater likelihood of their being caught—whether that is through fingerprints, DNA, increased policing or whatever.

Dr Wallace's submission refers to sexual offences. She mentions that a number of the offences would have been detected as a result of the perpetrators being known to the victims of the crime. Is it not the case that a number of cases do not get to court as a result of powerful DNA evidence, so the victim of the crime is not required to attend court and give evidence at the court session?

I am sorry, but I did not follow the question.

Your submission refers to victims of sexual offences. You claim that most of the victims are known to the perpetrators, so DNA evidence would not have been required. Do you accept that in a number of cases the victim of the crime is not required to attend court because the DNA evidence is so powerful that the perpetrator admits to being guilty in the first place?

That situation arises, but I do not think that it is relevant to the question of the extent to which the database is helping to convict people of those crimes. DNA evidence can be very important in convicting people of such crimes, but that is a matter of directly linking a known suspect to the crime using DNA evidence. We very much welcome that use of DNA and see it as being very important, but in such cases the database has not introduced a new suspect as a result of the retention of someone's DNA.

Fingerprints and DNA do not catch criminals and lead to convictions; they are part of the evidence that the police gather and present to the prosecutor to be presented in court. They are simply part of the armoury.

You talked about compelling information and you used figures from the Scottish Executive's consultation document on proposals to amend legislation on police retention of prints and samples, which says:

"As of 31 March 2005 it is estimated that there are around 198,000 DNA profiles on the Database which would previously have fallen to be removed."

However, the document goes on to say that about 190,000 of those profiles

"have not been linked with any subsequent crime scene. It could therefore be argued that the great majority of these profiles have been retained on the database for no purpose."

How many convictions have been secured as a result of matches with profiles that would have been removed under previous legislation?

I have no idea.

Perhaps that is because, as the Scottish Executive said in paragraph 3.2 of its paper, which immediately follows the paragraph that contains the figures that you quoted,

"It is not possible to ascertain how many convictions were secured because of the new law on DNA retention".

I refer you to the response that I gave to Mr Martin. Fingerprints and DNA do not in themselves lead to convictions.

Could it be argued that the balance that we have struck in Scotland is correct and that intelligence-led policing that uses forensics in the wider sense, including fingerprints and sampling, is better than a very large database that raises questions about how it is used for research? Such a database could be used for research that the SCRO might not know about, such as research into how likely people from certain ethnic minorities are to commit offences in particular postcode areas—that has been a research area. The trawling of a massive database can be expensive and bureaucratic. Would it be better to direct resources at carrying on with the intelligence-led policing system that we have developed in Scotland, of which the SCRO is a key component?

Intelligence-led policing is a UK-wide concept. As I am no longer a police officer, I will leave it for the police to describe and defend that approach to police business.

It is worth noting that the additional retained marks do not belong to people who are not guilty; they belong to people who have not been convicted in a court. Before people start to misinterpret that comment, I clarify that a person might have been dealt with through a fixed penalty or other non-court disposal. Children can be dealt with through the children's hearings system rather than through a criminal conviction, sometimes in cases of serious crimes and offences.

As I said, if there are more marks on the database and more samples or fingerprints that match the marks, the likelihood of a hit will be greater and the likelihood of detecting crimes or assisting the police in the detection of crimes will be greater.

That is why I asked why we do not sample every baby born in Scotland at birth.

I responded to your question.

You did not seem keen on the idea.

In fairness to the witness, Mr Purvis, you raise a matter for Government policy, which you should put to the minister. The witnesses might be able to contribute to the knowledge base.

I will pursue my colleague's line of questioning. Mr McLean talked about people who are not convicted in a court but who are dealt with through other means, such as a non-court disposal or the children's hearings system. Surely that is an argument for amending the system in relation to people who have been found guilty; it is not an argument for taking samples from everyone.

I am not making that argument. I am simply presenting my professional perspective. If we increase the number of marks in the database, we increase the likelihood of finding a match and detecting crimes and offences.

That is far from proved. We have discussed whether having more innocent people on a database increases the likelihood of achieving convictions.

It helps us. Not only will matches for fingerprints and DNA samples prove guilt; the absence of a match will prove innocence. If it is suggested that a fingerprint or DNA belongs to a certain person but it is proved that that is not the case, that person will then be excluded from police inquiries.

Will that not also increase the likelihood of mistakes being made?

I do not think so.

You do not think so. That is interesting—but there you go.

I will follow up with a couple of my own questions. Mr McLean, you spoke earlier about palm prints. You were the first person to raise that subject today. You also talked about other disposals that did not lead to the maintaining of information—either fingerprints or DNA—on a database. Is there a view within your service that you can share with us about the role or use of those disposals, for example in relation to a children's panel? Has that been a matter of discussion within your organisation?

I could give a response to that question, but I think that it would more properly be addressed to ACPOS.

That is fair enough. The matter crossed my mind and you just happened to be sitting there now.

Many young people under the age of 16 become involved in serious crimes but do not appear before courts. It is possible that the retention of fingerprints and other samples from them might help in the detection and prevention of other serious crimes.

Do you have a view on other disposals, Dr Wallace?

This is moving a little beyond my area of expertise, I am afraid.

You were speaking earlier about people in England and Wales who have been arrested and charged as opposed to merely getting arrested. Do you detect any movement to firm that up or broaden that in England and Wales?

The changes to the law that took place in England and Wales developed the other way round from here. As I understand it, DNA can be taken on arrest in Scotland but cannot be retained if the person is not charged or convicted. In England and Wales, the decision to retain was initially taken at a point when DNA was taken only on charge. The more recent change in the law, which came into force a year ago, involved taking DNA on arrest and retaining all the samples. That includes arrests for any recordable offence in England and Wales now, including begging, being drunk and disorderly or taking part in an illegal demonstration. It is a broad power to retain.

I take it from your written and oral evidence that your organisation has some concerns about that.

Yes, we do.

There are no further questions. I thank Dr Wallace and Mr McLean very much for making themselves available to us and for being so helpful in responding to our questions. I suggest that we take a short break while the minister and his team assemble.

Meeting suspended.

On resuming—

I welcome Hugh Henry, the Deputy Minister for Justice, who has come along to give us more evidence. He is accompanied by Ian Ferguson, from the bill team, and Carolyn Magill, from the office of the solicitor to the Scottish Executive.

Will you share with us the Executive's view on the aim of Paul Martin's amendment? Do you have a position on that?

We do not have a position on that as yet. We are listening with interest to the debate that is unfolding as a result of the amendment. I watched most of your proceedings this afternoon and picked up a number of useful points.

I took from the previous exchanges the clear view that two separate issues are involved, both of which need to be resolved. The first is a matter of principle: whether it is right or wrong to keep on record, anywhere, information on people who have not been convicted. That said, we probably do that already in other circumstances. The second issue, which has been raised a number of times this afternoon, is pragmatic: what can be justified in the solving of crimes.

What is your view on what the panel said about the collation and analysis of the figures from England and Wales of convictions that have resulted from the taking of additional samples? At this stage, there appears to be no numerical correlation between the two. Is the Executive looking at that? The issue was also raised in written submissions.

The point is a fair one, convener. I suspect that more work needs to be done on the subject. I am not sure whether the committee or, indeed, the Parliament would want to come to the conclusion that their decisions were based solely on statistical evidence. Other arguments, both for and against, also need to be considered carefully.

However, if weight is to be given to the proven value or otherwise of the figures, I suspect that we will need to dig a bit deeper into them. From the figures that the panel gave this afternoon, it seems that neither argument is sufficiently strong. Even if the increase in DNA detections did not follow on proportionately from the increase in the number of DNA records that are kept, there is an argument that we would be justified in keeping them if a number of significant crimes are solved as a result. We also need to give weight to the argument that, if there is no proportional increase in such solutions, there is hardly a justification for keeping the records. That said, we need to get much better information on the information and statistics—I too could understand them better.

I also hear the argument that detections have quadrupled through the use of DNA. I note that

"The annual number of direct DNA detections has more than doubled from 8,612 in 1999/2000 to 19,873 in 2004/05."

There is also the argument about how many of those detections led to convictions. The statistics also include cases in which the direct link to a DNA sample helped to eliminate someone from an inquiry. In terms of the arguments for and against the holding of such information, I suspect that we will need to delve much more deeply into the figures before we can put any significance on them. Issues other than those that relate to the statistics also need to be considered.

Before we get to stage 3 of the bill, perhaps you will share any information that the Executive has collated. Obviously, committee members and everyone from whom we took evidence today share the concern about the way in which figures are used. I will not quote the phrase about lies and statistics, but the issue is serious, as you rightly said.

I would be more than happy to share anything that we have. I would also be happy to have some further discussion—either formally on the record or informally off the record—about what the statistics may or may not mean. I may not understand all the statistics to the committee's satisfaction but, if my officials can help, they would be happy to do so. We are happy to help in any way we can.

I am grateful.

Aside from the fact that the evidence is possibly inconclusive at this point, what are the other benefits of retaining samples from people who have not been convicted?

Some argue for the retention of samples from people who have no convictions. There are probably three potential benefits of doing that.

The first is the early elimination of people who are not associated with a particular crime, which would relieve them of any continuing worry or anxiety that they may be associated with it. The second is the potential early identification of a suspect and, therefore, the potential early detection of the person who is responsible for a crime. The third, which I heard John McLean mention, is the potential deterrent effect: if someone knows that their DNA is on a database, they may think twice before they commit a crime.

I think of the terrible events associated with Soham. One of the criticisms that arose after the Soham murders concerned the inability of police forces in England and Wales to exchange information. Had they done that, more might have been known about Ian Huntly. We should reflect on the fact that Ian Huntly was never convicted of a crime until those terrible murders. We may ask whether information of any sort should be exchanged about someone who has not been convicted of any crime. I believe that Ian Huntly was charged in 2000 or 2001, but nothing happened thereafter. Under the rules that were in place at the time, his DNA could not be retained. One wonders whether, if it had been retained, the case could have been resolved much earlier and whether Ian Huntly might even have been deterred from committing the murders. We will never know. Members will have to consider such scenarios when they decide whether they are for or against the proposed amendment.

I want to pursue the issue of early elimination. I know that in cases of serious crime—sexual assault or murder—the police often invite people to give DNA on a voluntary basis, for the purpose of elimination. That is a well-known and positive practice. I know that in your consultation you were keen not to upset that arrangement, for obvious reasons. Under Paul Martin's proposed amendment, would the samples provided by such people be retained permanently? If they were, might that discourage people from coming forward?

Paul Martin's proposed amendment would not change the present situation at all. What was the second issue that you raised?

Do you think that the proposed amendment would cause anyone to hesitate before coming forward?

On a technical basis, it would not.

That is helpful.

I will probably fail to get this right again, but there is clearly a distinction between samples and profiles. We are talking about retaining both. There seems to be more concern about the retention of samples. I understand that profiles allow one to make a positive identification, so why do we need to retain both?

As Jackie Baillie indicates, there are two separate issues. The national database would retain only the profiles, whereas police in Scotland would retain the samples. It is possible that, when evidence is presented in court, having access to samples will be useful, as they can provide further information. However, we would not present the samples for storage on the national database. Only the profiles, which provide more limited information, would be stored there.

I am advised that the full samples would be used to confirm any DNA hits, as they are called. We could upgrade the database as technology improves. You will know that just this week there has been further discussion about how DNA can help solve crimes that go back years. Some crimes have been solved because our ability to detect what certain information means has become more sophisticated; we can now do things that we could not do when the sample was taken. The so-called hoaxer in the Yorkshire ripper case is one such example.

It is important to point out that in Scots law the DNA sample or evidence in and of itself is not sufficient, but it can help detection. In the case that we have heard about this week, a DNA sample that was taken a significant number of years ago led to the conviction. We believe that it is important to retain DNA samples of those who have been convicted, which is what happens now; only the profile would be stored on the database.

I am grateful that the minister said in response to the convener's question that he would be happy to share information about the efficacy of retention in respect of prosecution and conviction.

Some people have said that they are concerned that expanding the retention of samples could undermine public trust in police use of DNA evidence. Does the Executive have a view on that?

We do not have an official view. I can see the arguments for and against it.

Those who argue that expansion would undermine public confidence in police use of DNA evidence might be concerned that the information could be used improperly for purposes for which it was not intended. If we expanded the retention of samples, we would need to build in clear safeguards, checks and balances and prescriptive legal requirements about what such information could be used for.

Those who argue in favour of expansion might say that such fears are ill-founded and that if someone is innocent they have nothing to fear.

I have a question on Paul Martin's amendment, which, as I understand it, relates only to samples. What is the Executive's view on the retention of prints?

The same issue would arise. We are talking about retaining, from people who have not been convicted of anything, information that might help solve crimes. The principle is the same, regardless of which decision is made.

What is the annual cost of destroying DNA evidence?

It is somewhere in the region of £0.5 million.

If the Executive said that it opposed my amendment, how would that fit with the principle of best value, ensuring that the resources are targeted and that there are more police officers on the street? All the political parties represented here have made arguments about that.

Additional costs could be incurred if the police took DNA samples from more people. We should remember that under the current arrangements the police in Scotland do not routinely take DNA samples from everyone they arrest or everyone who subsequently goes through the court system. We are not saying that a DNA sample should be taken from everyone who is arrested. It is a matter for the discretion of the police. The legislation sets out the circumstances in which the police can take a DNA sample, if they believe that they need to do so.

Under the current arrangements, the costs to which Paul Martin refers are simply those of legally taking samples and sending them to the database through the appropriate channels so that they can be checked. Costs of about £500,000 are associated with the destruction of DNA samples from people whom the police thought should provide samples but who were not subsequently convicted. I hesitate to talk about best value in the context of this discussion, because we try to achieve best value in the current system and would continue to do so if the law were changed. Whether the £500,000 could be better used is for others to determine.

I asked Dr Wallace whether she thinks the retention of more samples on the database could prevent crime. Does the Executive think it could?

The argument has been made. I pondered whether the retention of information on the database would have prevented certain high-profile crimes. We can only hypothesise; God only knows whether those crimes would have been prevented. It is difficult to understand the mind of a criminal and whether a criminal always acts rationally is open to debate. Some people might think twice before committing a crime.

I was interested in the committee's discussion with the previous panel of witnesses. John McLean clearly supports Paul Martin's proposed approach, whereas Dr Wallace rightly considered the difference between principle and pragmatism. However, Dr Wallace was open-minded and could see the benefit of an approach that goes further than do the current arrangements. She said that there might sometimes be justification for the retention of samples from people who have not been convicted. We will reflect on the fact that GeneWatch UK thinks that there might sometimes be justification for such an approach.

We have been discussing the retention of DNA profiles, but are other details kept centrally by police or other authorities after someone has been arrested or questioned?

Yes. Personal details, for example names and addresses, would be retained. Coming back to the Bichard report, there was concern that there was insufficient exchange of information by police forces in England and Wales on a person who was not convicted. The police keep information about individuals and try to build up information about their behaviour patterns. As we have discussed in another context—indeed, Bill Butler posed the same question—that could lead to, for example, risk of sexual harm orders being placed on people who are not convicted. Such an order would not be taken out unless information was held about that person. Information is held on people who are not necessarily convicted of any crime, although that information might not be DNA.

I enjoyed the minister's Kenny Dalglish impression in relation to Paul Martin's amendment: "Mibbes aye, mibbes naw". It is a fascinating debate and there are balances to be struck between the various considerations. It is clear to us all that the contribution that DNA makes to solving crime is not at issue. The issue is the state keeping profiles of people who are not guilty of any crime. Might we just as well argue that DNA samples should be taken from the entire population? If that is ruled out, what distinction is there between the entire population and those who are not convicted of any crime?

I repeat the point that came up from Paul Martin's question. We keep information about people who have not been convicted of any crime because it is felt that doing so helps police intelligence. The police are able to build up profiles that can lead not only to crimes being solved but—if there is concern about people who have a particular pattern associated with them—to their prevention. We have reflected on that in relation to the use of civil legislation and the protection of people from potential sexual harm. I equivocate when it comes to Paul Martin's question simply because we do not as yet have a ministerial or Executive view. It would be wrong for me to come down one way or the other.

Colin Fox poses a more fundamental question. Should everyone's DNA be taken at birth? Given our present position, which is that we keep the DNA only of people who have been convicted, it would be a huge jump for the Executive to suggest, as Colin Fox does, that we take the DNA of everyone. Clearly, we are not minded to do so. There are issues of proportionality and effectiveness.

But how do the arguments relate to people who, at some point, have come into contact with the police, for whatever reason? As Dr Wallace says, there is a balance between preventing crime and protecting civil liberties. We are clear that we take civil liberties seriously. Whatever we do, we will build in safeguards to protect civil liberties as far as is humanly possible and within our obligations under the ECHR. Equally, we have a duty to protect the wider public from individuals who do not respect the civil liberties of others. That balance needs to be struck by the Executive, the committee and the Parliament.

I am not having a go at the minister—

I did not suggest that you were.

The minister's open-minded approach to criticisms that he has not heard before is perfectly laudable. However, in saying that DNA profiles are currently held of people who have not been convicted—

No, I said that we currently hold information on people who have not been convicted. We hold DNA samples only of those who have been convicted.

I simply want to point out, as Jackie Baillie did, that in Scotland DNA is currently gathered from people who give their consent. The written evidence from the Human Genetics Commission argues that the law on such matters should be standardised across Britain, but the commission's preference is that the law in England and Wales should fall into line with Scots law rather than the other way around. The commission argues that DNA samples should be taken only with consent and should be destroyed after a period of time. What is the minister's view of the evidence from the Human Genetics Commission?

Clearly, the commission is entitled to its opinion and it has made a worthy contribution to the debate. Indeed, I was struck by the way in which people of differing views made their views known in a measured manner earlier this afternoon. I did not detect extreme or ill-thought-out views in the very considered opinions that were offered. Although the validity and strength of the statistical evidence is clearly the subject of debate, Dr Wallace put the issue starkly for politicians. Politicians, rather than the commission or anyone else, need to come to a view on the right balance between preventing crime and protecting civil liberties. We need to do that within the broader debate that takes place about human rights. As members know, such decisions are not always easy.

Does the minister accept that holding information on individuals on the police intelligence database—which is managed by police forces in line with regularly reviewed procedures and protocols and which is under the name of the chief constable—is very different from holding DNA samples of those who have been arrested for being drunk or disorderly or for begging?

Is Jeremy Purvis asking whether people's DNA would be retained if they were arrested for begging or being drunk or disorderly?

As the minister rightly said, the question is about striking the right balance. He used as an example the fact that we currently hold information on people who have not been convicted of an offence. However, the conditions for entering that information into the police intelligence database are clearly set out. Indeed, the model that we operate in Scotland was recommended by Bichard for south of the border, so I suggest that he should reflect on whether he should have used the Soham case as an example of why DNA should be retained.

Nevertheless, decisions are taken about which individuals should be included in the Scottish database. That is very different from the rather indiscriminate approach in England and Wales that has allowed the DNA database to balloon to 3 million people. In Scotland, we have a targeted system under which decisions can be reviewed by the individuals whose DNA is retained and by prospective employers through Disclosure Scotland. The mechanism here is very different.

I will certainly reflect on that. Jeremy Purvis has raised the interesting point that individuals whose information is held have the right, under the protocols, to influence how that information is held. He raises a valid general point about the retention of information. At the moment, the Executive does not have a view on that, but the issue is worthy of some further thought.

If information is to be held about someone who has not been convicted, what checks and balances should be introduced? Dr Wallace also posed that question. Such a situation does not pertain in England and Wales, but given that the Executive does not have a view on whether we should move to having the system that operates in England and Wales or to having a system that lies between what we have now and what England and Wales have, we should reflect on what checks, balances, safeguards, scrutiny and protocols should be introduced if such a shift were to be made. Such measures should generally apply when sensitive information is held to ensure that it is not wrongly used. I am alive to the fact that there are always people who will seek to misused information, regardless of where it is held.

One of the checking mechanisms is review by the information commissioner. Are you aware that the information commissioner's office has come to the conclusion that the measures that Paul Martin's amendment advocates and those that are employed in the English and Welsh model are excessive?

I am aware that that is the information commissioner's view. I know that in England and Wales wider considerations were taken into account and that a policy decision was reached that has been tested in court and which I am sure will be tested again at European level. As was the case in England and Wales, the Parliament would need to reflect on such matters if it decided to agree to Paul Martin's amendment.

I have a question that follows on from Jackie Baillie's question, to which I do not know the answer. Whose property is the sample?

It would be the property of the police force that took it.

Would the police force own the sample and the profile?

Once the profile went to the database, it would be the property of the database. It would be difficult for individual police forces throughout the United Kingdom to own the information on the database collectively and jointly.

In effect, Scotland pays for each entry on the national DNA database. Is it the case that that information would be the property not of anyone in Scotland, but of the national DNA database?

It would be the property of the Scottish DNA database.

You said that the weeding of the information would cost £0.5 million. In the light of your reply to a parliamentary question that I asked about the individual cost of an entry on the national DNA database, if there was a large increase in the number of entries in the database, would that not simply mean that that cost would go up?

If you consult the Official Report, you will find that when I answered your question, I said that I could not comment on what the wider costs might be. We should remember that not everyone who is arrested will necessarily have their DNA taken. That does not happen at the moment and I have no reason to believe that it will happen in the future.

One of the reasons why the police do not routinely take samples for all offences—the examples of being drunk and disorderly and begging have been mentioned—is that we allow the police a degree of discretion when it comes to the detection and prevention of crime. You have highlighted three areas in which you say that the measures contained in Mr Martin's amendment might offer benefits—they might help with early elimination and early identification and could act as a deterrent. Would those benefits not be undermined if the police were not forced to take DNA samples from everyone who was arrested?

I gave an answer to a specific question about what the potential benefits might be; I was not advancing particular views that I hold.

Forcing the police to take samples would be an extreme measure. It would be perverse of those who oppose any extension to the retention of DNA samples to suggest that the logical conclusion is to go even further and take a sample from everybody. By leaving the matter to the police, we allow them to identify potential crimes for which it might be worth taking DNA samples and crosschecking, for whatever reason.

We have an open mind about whether Paul Martin's amendment should be supported, but I would need to be persuaded that it would lead automatically to the police taking DNA samples in every case, whether or not the offence was trivial and whether or not the sample would have wider application now or in the future.

You said earlier that more analysis might be required of some of the information from the Home Office and GeneWatch UK that has been cited. Do you agree that, as the Executive's consultation paper says,

"It is not possible to ascertain how many convictions were secured because of the new law on DNA retention"

in England and Wales? No further analysis can be done of the number of convictions that have resulted from retaining DNA, because

"it is impossible to know for certain why a jury makes a decision in any particular case",

as the consultation paper says.

There is no point in doing further analysis of convictions, although we are all interested in that. As you know, the clear-up rate in Scotland is considerably higher than that in England and Wales, which we are pleased about.

Different issues are involved. I accept that much of what we are discussing is relatively new, so it will take time to know the impact. When I spoke about having to investigate the statistics further, I gave the caveat that other matters—not simply statistics—will influence any decision. When the Executive reaches a view for or against, that will not be predicated on statistics alone, although they will have an influence.

To ask whether DNA evidence was the reason why a jury or a judge chose to convict does not pose the question in the proper context. In Scotland, DNA evidence on its own would not be sufficient; corroboration would be needed. DNA evidence might provide a lead—that happens at present when someone is on the database because they have a conviction. However, whether the DNA evidence would lead to a conviction is a moot point, as other factors would need to be established. The strongest factor that led to a conviction would be determined by whoever made the decision to convict. It would be wrong of me to say that DNA evidence is a determining factor one way or the other.

All sorts of figures have been mentioned, both today and previously, but we need to get behind them and understand as best we can the significance of the increase in DNA detections from 6,151 in 1998-99 to 20,489 in 2003-04 and the increase in the percentage of matches from 29 per cent in 1998-99 to 45 per cent in 2003-04. If one is to give any weight to a statistical argument, one needs to understand the significance of the statistics. I want to consider the wider issues. The statistics might have an influence, but it would be wrong to say that they are the sole determinant.

It is the—

Very briefly.

I will leave it there, convener.

Thank you. Maureen Macmillan has a supplementary question on that point.

It is actually on something that the minister mentioned in his response to an earlier question. You discussed who would have their DNA kept and who would not. GeneWatch UK is against the retention of the DNA of people who have not been convicted, although it said that it would countenance that in some cases, such as those involving sexual crimes. How easy would it be to state in legislation that DNA retention is allowed in some cases in which people have not been convicted but not in others?

To some extent, that question takes us back to the debate that we had on risk of sexual harm orders, which posed difficulties of principle, never mind difficulties with the framing of legislation. In the decisions that the Parliament has made in recent years, the protection of children has been high on our agenda. We have had to make some difficult decisions that would probably not have been contemplated years ago.

I think that Dr Wallace was just reflecting the public's anxiety about the retention of DNA in certain circumstances. She stated GeneWatch UK's view that there should be restrictions. I am not sure how we would go beyond our current position to include not only those with a conviction but cases in which there is no conviction but potentially a sexual element. However, I am clear that a different amendment would be needed from the one that Paul Martin has lodged. I am not sure whether we could construct an amendment that dealt with solely sexual, but not violent, crimes, or an amendment that dealt with certain types of sexual crimes. However, Dr Wallace's contribution reflected a recognition that a justification could be made for going beyond the current position.

I have a rather delayed supplementary question. In response to a question from Jackie Baillie, the minister gave three reasons why people might support the retention of DNA samples. One of those reasons was that it could lead to the early elimination of a suspect, which would reduce their anxiety levels and be much easier on them. However, the current situation is that a DNA sample can be taken from a suspect and they can be eliminated from the inquiry. It is not easy to match up the argument about retention and the argument about elimination. We use DNA samples to eliminate suspects at present.

We can certainly eliminate them if they give a DNA sample voluntarily, but if they do not they have to go through the process of being arrested and questioned. If the information is there, it might not get to that stage. They might not have to go through the stress and strain of being taken in for questioning.

Surely that is their decision. If they do not want to give a sample voluntarily, that is up to them. If they decide not to give one, they know the likely outcome.

No. Sorry—I am probably not explaining myself—[Interruption.]

Let us pause while Ms Baillie manages to sort out her mobile phone, which she forgot to turn off.

Somebody is making sure that Jackie Baillie is on message—at least it is not me who is getting a message.

It takes more than a mobile phone, minister.

It is possible that I did not explain myself well. At the moment, people can give DNA samples voluntarily, which could lead to someone being identified—either the individual who gives the sample or a relative.

In a recent horrific murder case in Glasgow, the person responsible was identified because, by a quirk, a relative gave a DNA sample voluntarily. That led to the identification of the person responsible for the crime. However, someone could be arrested and taken in for questioning in a particular set of circumstances—I do not have to hand the form of words that describes the circumstances, but it is something like "for a crime that could lead to a conviction"; I can get the correct wording to the committee. In those circumstances, safeguards exist that would justify the police taking a DNA sample.

Stewart Maxwell asked whether taking a sample could lead to someone being eliminated from inquiries. If a person's DNA were held, it is conceivable that a check could be made and that that person could be eliminated from inquiries without their having to be brought in, questioned or charged.

I accept what you are saying and understand the differentiation that you make. Interestingly enough, however, the example that you gave was of a successful police case under the current system.

The reason why I gave that example was because you posed the scenario of someone giving their DNA voluntarily—some people do. I suppose that it is a moot point whether the person who was eventually found guilty of that murder had been arrested previously for something—I do not know their background and have no reason to know it. It is a moot point whether their record might have been on the database had things not been as they are just now. Who knows? The crime might have been solved earlier, but I have no way of knowing.

I will move on to a separate point. Paul Martin asked whether having a large number of DNA profiles on the database would lead certain people to think twice about committing a crime in the future. In other words, it might act as some sort of crime prevention measure. You touched on the Soham case as a hypothetical example of that. I am not aware of any evidence of claims that support the crime prevention argument—I accept that there are such claims, but I know of no evidence that databases actually prevent crime.

Are you aware, as I am, of lots of research that shows that criminals do not think ahead? Many studies from all round the world show it to be the case that criminals think no more than five minutes ahead and have no concept of the consequences of their actions. One of the interesting aspects of research on criminality is that people who are involved in crime tend to lack the foresight that people in the general population have. People in the general population think about the likely consequences and outcomes of their actions. Are you aware of the research that shows that criminals do not look ahead? Do you have any evidence that contradicts that research and shows that a DNA database would act as a crime prevention measure?

If you check the Official Report for my earlier answer, you will find that I did not state unequivocally that people would be deterred by such a database. I do not know how the criminal mind works. I acknowledged that some people act first and think later.

Perhaps I can help you, minister. You were referring to the evidence of Mr McLean from the SCRO—the suggestion was not yours originally.

Thank you, convener—I was just about to say that John McLean made the comment. It is helpful to put the matter in context. I can comment, but the committee would need to dig further, because it is not for me to suggest that such evidence is crucial. Some people certainly commit crimes without thought. Some of them live to regret it, but others do not care. Equally, some people plan crimes and take significant steps to hide their identity, as with the recent big bank robberies in England and Northern Ireland. People who act first and do not think might not try to disguise their identity by wearing wigs, glasses, gloves or balaclavas, but others think through the consequences of getting caught and the information that may be left behind. With all due respect, the question is probably better addressed to the police than to me.

The Executive will be aware of the concern about the current system. Some people think that there is insufficient oversight of the system and are not convinced that sufficient safeguards are in place to ensure that no misuse occurs of the information that is retained on databases. What is your view of those concerns?

We have clear rules, legal expectations and protocols. I hope that the safeguards are robust and effective. However, it would be complacent to suggest that the system cannot be improved. If genuine concerns arise and if people have ideas about how the present system, or any future system, could be better regulated or governed to give the public the safeguards that they want, it is incumbent on all of us to consider them—they would be worthy of discussion.

Are the safeguards that are currently in place monitored?

Not that I am aware of. The various agencies may well scrutinise the safeguards, but I have no knowledge of that.

As this seems to be my afternoon for asking technical questions, I have another one, so bear with me. There may be a discrepancy in existing legislation on the destruction of DNA samples. The Criminal Procedure (Scotland) Act 1995 sets out the powers for destruction of samples in section 18(3) and defines samples in sections 18(2) and 18(6). I am reliably informed that section 55 of the Criminal Justice (Scotland) Act 2003 introduced section 18(6A) into the 1995 act. We are not sure whether section 18(3), which sets out the powers for destruction, covers section 18(6A). I just wanted to share that technical information with you. Perhaps we can have a statement on whether section 18(6A) is covered.

We are aware of the problem and are working on a resolution to it. To an extent, our reaction will depend on what the Parliament does with Paul Martin's amendment on the subject. If we are talking about the same section, Paul Martin's amendment would address the problem. If Parliament decides not to agree to the amendment, we will have to find another resolution, because the 2003 act did not intend to create that problem. We will have to liaise with the committee on that through stage 2 and up to stage 3.

Aside from fixing that anomaly, depending on whether the committee supports Paul Martin's amendment, does the Executive have any legislative plans on the retention of prints or DNA samples?

We are still reflecting on Paul Martin's proposal. We have not decided to lodge any amendment.

I appreciate that we have run slightly over time, but we are all in the same boat. Paul Martin's amendment must be dealt with somehow and we are grateful that you were able to come and share your thoughts with us prior to the debate on the amendment.