I beg to move amendment No. 351, in

clause 65, page 40, line 5, leave out 'and compelling' and insert ', reliable and substantial'.

With this we may consider amendment No. 352, in

clause 65, page 40, line 9, leave out subsection (3).

Let me make it clear that I strongly support the proposal to give the Court of Appeal powers to reopen cases in which persuasive new evidence comes to light. I tabled the amendments because I have doubts about the wording that is used with regard to the process of determining which matters should come before the Court of Appeal. My concern is twofold. First, the threshold set out in subsections (1) to (3) is too high, and secondly, because of the wording, there is a danger that there will be a determination—or something close to a final determination—of any case that comes before the Court of Appeal before it has had the chance to go back before a jury to be considered in full.

A sifting process is already in place, through both the Director of Public Prosecutions and the Court of Appeal. The DPP will bring the matter before the Court of Appeal only when he has made a determination. The amendment would require new, reliable and persuasive evidence to be put before the court before the matter could be referred back for trial by jury. That seems to me to be the right approach.

I am concerned about the words

''highly probably that the person is guilty of the offence.''

If someone has been found by the Court of Appeal to be highly probably guilty, that creates difficulties for a jury that is aware of that finding in making its subsequent determination. There might be some resistance by the jury to the initial finding. I am

object not to the principle but to the wording. As I mentioned last week, the Court of Appeal can currently find a conviction unsafe and unsatisfactory and refer it back for a retrial. My objection is not to the principle but to the fact the clause as drafted is not helpful. My particular concern is that it is not clear that the clause provides for the defendant to be fully represented.

Clause 67(4) states that the defendant can be present at the hearing. I wish to hear more about what exactly that means. Does it mean that the defendant will be represented at the hearing and will have the right to make representations? I would be grateful to hear from the Minister on those concerns.

I welcome the amendments, because the hon. Member for Wrexham (Ian Lucas) has gone to the heart of the issues that arise. If I do not entirely agree with him—although I shall be interested to hear what the Minister has to say about amendment No. 351—I have a great deal of sympathy with the points that he has made about amendment No. 352.

As we shall see later from the amendments that I and my hon. Friends have tabled, the wording of the clause—on the findings that the Court of Appeal must make—gives me exactly the same sense of unease as it gave the hon. Gentleman: namely, that a finding of high probability that the defendant is guilty of an offence appears to be a prejudgment of the decision that the jury must make at the retrial.

I also agree with the hon. Gentleman that, ultimately, the problem may be simply a matter of wording. I find it hard to believe—and I hope that the Minister will respond positively to our amendments later—that a form of words cannot be found that indicates the high test with which the Court of Appeal must be satisfied before it allows a retrial but that does not give the impression that the court has formed a view.

That is of more than academic interest, because although there will be reporting restrictions, it is likely that people who are aware that there has been an application for retrial will examine the statute to see what the test is that the Court of Appeal has to be satisfied with. We shall consider later whether we may be able to prevent the test from being reproduced, but if it is reproduced, the Court of Appeal will appear to be prejudging the issues that the jury must consider.

I would expect the Court of Appeal to make its decision on the basis of the strength of the prosecution case. The court will not have to consider what the defence may choose to present at trial for the jury to consider, except in so far as it may have received representations about the degree of unfairness that may apply to the defendant, in terms of defences that he may originally have wished to make, or matters that he may have wished to put before the court.

I find the wording unfortunate. We must be capable of doing something better with it. The clause appears to be a clear prejudgment of the issues that must be considered at retrial. I shall return to that in greater detail on my amendments, although I fully

acknowledge that the hon. Gentleman has been trying to tackle the same issue.

The hon. Gentleman raises an interesting point in amendment No. 351. What does ''compelling'' mean? ''Reliable and substantial'' may be a better test, especially as that appears to be the test to say that evidence is compelling in subsection (3), albeit that there it is taken with the words,

''when it is considered in the context of the outstanding issues, it is highly probable that the person is guilty of the offence.''

I would be unhappy about getting rid of subsection (3) entirely, as it provides an extra layer of protection for the defendant. Evidence should not only be reliable and substantial but should be taken in context. I would not want the context aspect to disappear. That said, I would prefer to retain the word ''compelling''—although there might be some better alternatives—and for subsection (3)(c) to be altered so as not to give the impression that the jury's decision has been pre-empted.

I am grateful to the hon. Gentleman, and pleased that the Committee has widely registered the fact that there are problems with the current wording.

My hon. Friend is touching on the nub of the problem. On prejudging the jury's decision, does he think that the jury's view might be affected if it was widely known among potential jurors that the test under subsection (3)(c) had been applied to the case and had been passed?

I agree entirely. We shall have to look at how the reporting restrictions will operate. Unless I have misunderstood the thrust of the legislation, I think that they will be far more complicated and difficult to operate than I had previously thought. In particular, there is an issue to do with the extent to which those restrictions apply in Scotland, on which the Minister might be able to reassure us.

It is not beyond the wit of jurors to become aware of the test by which the Court of Appeal sends a case for retrial. The plain meaning of the clause as currently drafted appears virtually to be an indication that conviction is inevitable. When we come to consider the word ''scientific'' we should be wary of an assumption that scientific or other evidence will be the clincher or provide foregone conclusions—juries should certainly avoid those—in such cases. I would assume that the basis for retrial is ''reliable and substantial'' evidence, as the hon. Member for Wrexham said, taken in context. We shall come on to what might be an alternative wording—I do not want to repeat myself later, so I shall hold fire on the matter. However, there must be some alternative phraseology that does not use the words

''highly probable that the person is guilty of the offence.''

The words ''might be guilty'' might be better than ''is guilty'', although I wait to hear from the Minister. This is not merely an exercise in words and semantics—we must get the clause right.

The Committee owes a debt of gratitude to the hon.

Member for Wrexham for the amendments. As the hon. Member for Beaconsfield said, they strike at the heart of the principal objection to the provisions. I speak as a member of the league of non-lawyers on the Committee—I think that we are a minority—and even I can see that the provisions raise serious questions about the conduct of a subsequent trial.

The great paradox is that the higher the test for a retrial to take place, the more hurdles the prosecution must jump to bring a case back and the stronger the indication to the jury that the defendant is guilty. Under the present arrangements, any juror who is aware of the process will know that the prosecutor has not only determined that the defendant is guilty but convinced the Director of Public Prosecutions—in person—that the evidence overwhelmingly suggests guilt.

That will be par for the course for most criminal trials, but there will also be the test applied by the Court of Appeal. Under the clause, the court will be asked to consider whether the evidence suggests that it is

''highly probable that the person is guilty of the offence.''

If the court holds that to be true, its decision is likely—even with the best will in the world and the best possible directions from the trial judge—to influence jurors before they have had a chance to hear the evidence. That will happen even if there is no reporting on the case, although, as we know, outside influences will be brought to bear on juries in such high-profile cases.

I have some sympathy with the view almost expressed by the hon. Member for Wrexham, and a formulation analogous to that for unsafe or unsatisfactory convictions might be a better way forward. As the Bill stands, the Court of Appeal may determine that there is a high probability that a person is guilty.

My objection to what the hon. Gentleman said is the same as that of the hon. Member for Beaconsfield. By rejigging the clause and introducing the phrase ''reliable and substantial'', he would, of course, remove the problem of the high probability of someone being guilty. However, he would also remove the context of the outstanding issues. Any evidence that is adduced may be reliable and substantial but have nothing to do with the terms under which the person was acquitted in the first instance. The problem with the hon. Gentleman's formulation is that the evidence might be irrelevant to the process.

We must get the Bill into a workable form that does not fatally prejudice proceedings, although, as we have heard, very few cases will be handled using this process, because of the difficulties involved. The worst outcome will be if we raise people's expectations that the system is capable of delivering a fair trial, only for them to find that it is not. We must ensure that the wording of the clause works against that outcome.

I hope that the Minister will give the matter a great deal of thought, and we have time to consider alternatives. I do not ask him to reduce the legal

hurdles needed for a retrial but to find a way of formulating the provisions so that they do not innately prejudice the jury that is called to hear the trial. Under the present terms of the Bill, however, they inevitably do so.

I share the concerns that have been expressed. As I have said, however, I totally support the principle behind the proposed change in the law. It is important, however, to get the details right. I appreciate that a difficult balancing exercise is required. We must not set the test too low, creating concerns that the police will not investigate a case properly the first time and that people will be prosecuted again and again. On the other hand, we must not set the test too high or too wide. As the hon. Member for North Down (Lady Hermon) powerfully highlighted several times last week, it is possible that the test may predispose a jury to convict.

The clause should establish a very high test regarding the calibre of the fresh evidence and its likely consequences in a trial. Surely, the job of the Court of Appeal is to determine whether there is enough evidence to justify quashing an acquittal and ordering a fresh jury to look at the matter, in which case it should focus totally on the quality of the evidence. Perhaps phraseology such as, ''the evidence must be new and compelling, and it must be essential, in the interests of justice, that a new trial should take place'' would be appropriate. Perhaps that would sufficiently express how powerful the new evidence would need to be—and deter the police from ever thinking that they could jump this hurdle as a result of negligence or neglect—without making any presumptions about guilt or innocence.

Will the hon. and learned Lady say whether she believes that the words,

''highly probable that the person is guilty of the offence''

replace the presumption of innocence, which is guaranteed by article 7 of the European convention on human rights, with a presumption of guilt before the retrial?

The judge should have a duty to restore the presumption of innocence, if there is a gap between what might appear to be so and the actual situation. He could do that by way of a direction to the jury that negates any potential prejudice resulting from the decision of the Court of Appeal. I have considered what sort of direction that might be, but it is difficult to think of an appropriate direction on the basis of the contents of the clause.

Before we move on to consider how the provision might predispose a jury in a certain direction, I want to say something about the impact of the extra wording. In the context of the Court of Appeal, I believe that the requirement is over-wide and unnecessary. It would take the Court of Appeal beyond its function, which is solely to decide whether an acquittal should be quashed and a case considered by a fresh jury. That decision should be taken on the basis of the calibre of the evidence.

It is extremely difficult to know how to protect a defendant against an assertion by the best-known and highest criminal court in the land that it is highly probable that he is guilty. It is almost certain that Lord Woolf, the Lord Chief Justice, will decide the first six or eight cases. He has interfered in all sorts of political issues in the past few days, but that is a separate issue. It is well known that he is the highest judge in the land. By the time that a judge reaches that sort of age and standing, he has about 40 years' experience in the court system and his judgments must be followed, and are followed, by every other court in the land. If such a judge thinks that it is highly probably that a person is guilty, that will be a huge factor and is likely to predispose a jury.

That additional requirement is unnecessary for the decision that the judge would have to take, so let us consider how to restore the presumption of innocence in the face of such prejudice. I simply cannot imagine what a judge could say. If the Court of Appeal had considered the calibre of the evidence and said that it was so new and so compelling that a retrial was essential, the judge could, and would, then say to the jury, ''Their Lordships have rightly deemed a retrial essential, but they have considered only one piece of new evidence and have had no opportunity to consider the defence of this individual. That duty now falls to you.'' In such a case, the jury should accept that the decision is theirs and that there is enough evidence in the interests of justice to proceed to a new trial.

How can one devise a direction along the lines of ''Although their Lordships have said that this man is highly probably guilty, we must start afresh and disregard that''? We must bear in mind the impetus behind the Court of Appeal's adjudication.

Does the hon. and learned Lady conclude from that argument that she would be satisfied if there were tests to define the evidence but that there should be no statutory reference to what conclusion such evidence may lead to in the Court of Appeal? Does she conclude that the only matter that the Court of Appeal should decide is the nature of the evidence and that it should not reach a conclusion on guilt or innocence?

Yes. That is its function. There is an alternative way of looking at the argument: all the Court of Appeal should do is know the case sufficiently well to put the fresh evidence in context, but it should not know about the defence or know in detail what evidence is likely to be called for the defence. However, if we require the Court of Appeal to say that it is ''highly probable'' that a defendant is guilty, it may shrink from doing so. It may think that that is simply outwith anything that it can express a valid opinion on. That would be a dreadful shame, as it would spoil the whole affect of this legislation. I invite the Government to do some further thinking on this.

I support the preceding contributions, from my hon. Friend the Member for Beaconsfield, the hon. Member for Somerton and Frome (Mr. Heath) and the hon. and learned Member for Redcar (Vera Baird). It behoves the Minister to

listen to the broadly similar points that were made by a coalition of lawyers and non-lawyers, which is a happy union in the Committee. The hon. and learned Member for Redcar is right: it asks a great deal of the Court of Appeal to decide whether it is ''highly probable'' that a person is guilty on the basis of one piece of evidence.

I speak as one who broadly supports what the Government are trying to do but who doubts whether they have got this part of the mechanism absolutely right. There should be a stiff test to be passed before retrial, but such a test should not deprive the defendant of the opportunity of a fair trial. The hon. Member for North Down was right in her intervention: this does seem to disturb the presumption of innocence. It could also be highly prejudicial if it were known to a jury. The hon. and learned Member for Redcar is also correct to draw attention to the fact that such cases will inevitably be very high-profile, and there is a chance that they may be decided by Lord Woolf. There will be huge public interest in them, particularly in the first of such cases to be decided. What becomes known to the public when the case is being reported will be very important.

I shudder to think of a television reporter outside the royal courts of justice reporting that in such-and-such a very important case the Court of Appeal has just decided that, according to subsection (3)(c),

''it is highly probable that the person is guilty of the offence.''

The Minister must consider such a possibility, given the huge public interest that such cases will generate. He would do well to take on board the concern and look closely at the mechanics.

I rise again because I am conscious that we shall soon come to a group of amendments—amendments Nos. 305, 306, 307, 265 and 311—that will not need to be moved if we have properly completed the present debate. The more I look at it, the more convinced I am that those should properly be included in the present group. The Committee will see that I and my hon. Friends seek in that group of amendments to amend to detail of the text. For instance, we would remove the word ''is'' and put in ''appears to be'' guilty. Also, instead of ''the person is'', we would have the clause state that the

''jury would find the defendant guilty''.

Those suggestions were proposed for the purposes of discussion, but it is clear that they are subsumed in the present debate.

The longer the debate, the more I have come to realise that it is not easy to tinker with the wording of the clause in the way that I suggest—nor, I am bound to say, as the hon. Member for Wrexham has tried to do. I share the view of the hon. and learned Member for Redcar that the Government need to go away and think the clause through afresh. Her test, that the clause should be reworked to remove the business of probabilities so that new and compelling evidence should result in a retrial, seems a much better way of expressing what the Court of Appeal would have to do. The downside of our proposals is that they still

preserve the probability issue being determined by the Court of Appeal, which is the very thing that we need to get away from.

I hope that we may have saved time, Mr. Illsley, and I am grateful for your indulgence in allowing me to speak again, but I do not want to start a further debate on an issue that, in my view, will have been exhausted if the Minister can give a positive response.

I hope that I can help, too, by not having to speak to amendments in a later group. As my hon. Friend the Member for Somerton and Frome said, our view is clearly along the same lines. However, I anticipate those amendments; we would certainly prefer the approach argued for in previous debates by the hon. Member for North Down and by us, which was put today by the hon. and learned Member for Redcar: that the Court of Appeal should not come to conclusions but should only judge the quality of the evidence. That would clearly be better, and it would parallel what the Court of Appeal does now when asked to consider possible wrongful convictions. It does not usually come to a view about such cases; it assesses the evidence and then remits it to be considered by another court, although it sometimes takes that decision as part of its deliberations.

If the Minister was willing, as I hope he will be, to take away that part of the clause, we could make significant progress. Irrespective of our views on what sort of evidence should trigger retrials, we might then be able to come up with something that was far less of a worry to the Committee and those outside.

This has been an important debate, and I thank my hon. Friend the Member for Wrexham for initiating it. The hon. Member for Beaconsfield was right to say that my hon. Friend got to the heart of the matter in relation to not only this group of amendments, but subsequent ones, and I hope that I can enable the Committee to make the speedy progress that we all desire.

We recognise that we need to get the balance right between where the test is pitched in relation to permitting a retrial and the concerns expressed in previous debates. For example, let us suppose that someone were to argue for treating the matter just like any other. In the normal course of events, the Crown Prosecution Service would decide, having taken another look, just to proceed, and all the people about whom the hon. Member for Southwark, North and Bermondsey (Simon Hughes) expressed concern would think that they might be touched on the shoulder at any moment. Committee members of all parties have recognised the need to set a high threshold for testing the new and compelling evidence, and it is important to focus on that now.

Let me say in passing, with reference to the point made in an intervention by the hon. Member for North Down, that there is no intention to jeopardise or undermine the fundamental principle of presumption of innocence, which is still applied when a retrial starts. We also certainly intend that the defendant would be represented at the Court of

Appeal, and able to make representations. I accept entirely that that should happen.

My hon. and learned Friend the Member for Redcar was right to remind us that in considering clause 65(3), and in particular paragraph (c), we are thinking about what the Court of Appeal should take into account in coming to a judgment on the nature of the evidence. It is not a matter of substituting its verdict for that of a jury if there is a retrial. The Court of Appeal would be examining the evidence, but not being asked to deliver a verdict. That entails a different test.

A difficulty exists, and we must be honest about it. The new evidence should be relevant to the acquittal and it should show, using a pretty high test, that the first acquittal was not right or was unsafe, however one wants to put it. That is an important difference. In essence, an extra stage is being proposed, in addition to what happens in all cases where the CPS considers the evidence.

Every day of the week, the CPS examines evidence and reaches the view that it is highly probable that the person concerned committed the offence, or even that he is bang to rights. It then applies the two tests: first, whether there is a 50 per cent. chance of conviction, which will clearly be heavily influenced by the nature of the evidence, and, secondly, the public interest test. The Court of Appeal would be undertaking an analogous function in respect of the particular and unusual circumstance provided for in part 10. Those are the matters that we are evaluating.

I am persuaded, however, that there is a need to re-examine the provision. A reason for that was touched on by one or two Members—it could be argued that there is an apparent conflict between clause 65(1), with its reference to evidence that

''the acquitted person is guilty'',

and clause 65(3)(c), which states that for evidence to be treated as compelling it should be

''highly probable that the person is guilty of the offence''.

For that reason alone, I intended, before the debate, to undertake further to consider and reflect on the matter. I also give the Committee an undertaking to reflect on the points that all Members have made.

I sense that there is no disagreement about what we want to achieve. I understand the argument that clause 65(1) should not be worded so as to appear to prejudice fair consideration of the case on retrial, although I thought that my hon. and learned Friend the Member for Redcar did a pretty good job, towards the end of her remarks, of setting out the guidance and direction that a judge would give a jury in a retrial in asking its members to put all other considerations out of their mind and to examine all the evidence.

It is important to bear in mind the fact that a decision to order a retrial will be based not only on the new and compelling evidence, but on all the evidence. The jurors would be asked to listen to it and reach a judgment. The two protections that we should hold on to are, first, that the trial judge will be responsible for proper direction of the jury and for ensuring that the trial is fair and, secondly, that we have to trust the jury

to reach a fair verdict on the basis of the evidence that it will be asked to consider at a retrial.

The ultimate protection for the individual and the ultimate answer to the question asked by the hon. Member for North Down on whether the trial can be fair is that we have to trust the jury. I undertake to reconsider the precise wording of the clause, because important points have been raised. No doubt we shall return to the matter.

I am grateful to the Minister. If the Government think again, they might find matters easier. Indeed, they could almost start again on the basis outlined by the hon. and learned Member for Redcar.

The Minister's comments show why the clause has been drafted as it has. He used the expression the ''acquittal is not safe'', but that cannot have anything to do with the matter, as an acquittal is never unsafe. To take that approach is to consider the clause's intentions from the wrong angle. Adopting the concept of the acquittal being unsafe creates the type of test that produced the clause as drafted. That is why I like the simplicity of the approach of the hon. and learned Lady, which gets round the issue by saying, ''Here is the evidence, and it is in the public interest that, notwithstanding all the normal protections under double jeopardy, there should be a retrial and the jury can make up its mind.''

In my experience of Committees considering legal drafting, those who draft tend to find a precedent, just as lawyers, when drafting pleadings, always look for guidance in the previous pleading that fits the case. I fancy that the same approach has been adopted here. There are processes by which we can overturn unsafe convictions, but how do we overturn unsafe acquittals? We are dealing with a different animal.

The hon. Gentleman should not read too much into the words that I have used. I was reflecting on the comment of the hon. Member for Somerton and Frome about unsafe convictions. I grant entirely that the circumstances are different, but I was not trying to indicate that that is necessarily the direction of our thinking so far as the wording is concerned. I was struck by the point that, in circumstances in which a conviction is deemed unsafe, it can be agreed, in the interests of justice, that it should be reconsidered—no more than that.

I am not tying the Minister down to a form of words. However, in trying to understand how the clause emerged as it did, I imagine that there was a deep desire to ensure that the test was a high one. As an inevitable consequence, the jury's later decision making has been pre-empted. That is the mischief that needs to be addressed, even if it has occurred unintentionally.

As the Minister pointed out, the consideration of the Court of Appeal, even though it will have power to call witnesses and do all sorts of things, is bound to be partial. The trial process, particularly the defence case in response to any new evidence that emerges, will not have been tested and scrutinised by the Court of Appeal, which will take a decision based on the new, compelling, reliable and substantial evidence advanced

by the prosecution on the basis that it brings a completely different dimension to the case against the defendant. There will not be an opportunity for the defendant's response to that—except, perhaps, some preliminary comments—to be fully considered. That provides the Minister with his let out, and the drift as to how the clause might be reworded.

It is worth repeating that merely tinkering with a few words may be unproductive and may not achieve the intended result. Obviously, the less the wording has to be changed, the better, but the more I look at my amendments, the less I think that they meet the nub of the issue any more than the Government's original draft does.

I concur, although the Minister helpfully said that he would consider the issue. I therefore want to implant in his mind the critical factor, which the hon. and learned Member for Redcar expressed so well. It is not for the Court of Appeal to determine the defendant's guilt, but for the jury in the subsequent trial. It is difficult to sustain the idea that we should include phrasing implying that the Court of Appeal may determine guilt on a partial review of the evidence—that would inevitably be its lot in the process—or that it would not prejudice the jury's view by transmitting its decision to the subsequent trial.

I hope that the Minister principally addresses the point that it is in the interests of justice and in the public interest for the jury to consider the evidence in the round, and that includes any new evidence, which must be reliable and substantial, as we shall discuss. It is not for the Court of Appeal to determine explicitly, as the clause requires, the probability of guilt on the part of the defendant. If the Minister accepts that principle, he will start the process of considering the rewording of the clause on the right basis. I hope he confirms that that is his view.

I am grateful to my hon. Friend the Minister for his assurances, and I beg to ask leave to withdraw the amendment.

Amendment, by leave, withdrawn.

I beg to move amendment No.334, in

clause 65, page 40, line 5, after 'compelling', insert 'scientific'.

With this it will be convenient to discuss the following amendments:

No. 335, in

clause 65, page 40, line 7, at beginning insert 'Scientific'.

No. 336, in

clause 65, page 40, line 9, at beginning insert 'Scientific'.

No. 271, in

clause 65, page 40, line 9, after 'if', insert

'it is scientific in nature and'.

No. 337, in

clause 65, page 40, line 24, at end add—

'(8) In this Part ''scientific evidence'' means evidence derived from the analysis of DNA'.

No. 338, in

clause 66, page 40, line 33, after 'new', insert 'scientific'.

No. 339, in

clause 66, page 40, line 38, after 'new', insert 'scientific'.

Amendment No. 334 stands in my name and that of my hon. Friend the Member for Somerton and Frome, and is supported by the hon. Member for Beaconsfield and his hon. Friends. The other amendments are similarly supported, although amendment No. 271, which has a similar goal, was tabled by the hon. Member for Beaconsfield and his hon. Friends.

Our amendments argue that the new evidence that triggers the processes that we have discussed should be limited to scientific matters, and that has two obvious implications. First, we are addressing the fact that the police see advances in science, particularly as regards DNA, as the most significant development in the investigation of crime and one that may help to clear up a significant number of cases. The other side of the coin is that it may be wise, if we are thinking of removing the rule against double jeopardy and allowing retrials, to limit to scientific issues the evidence that may allow retrials to occur. However, that would preclude the other reasons why evidence might come up, such as—in particular and most worryingly—the police and those investigating matters not doing their job thoroughly the first time.

The greatest concern among the public, in particular the informed public who contemplate such matters, is that if we allow retrials after people have been found not guilty, there will be an incentive for the police to have a second go when they did not get there the first time. That may countenance or condone less than satisfactory investigation the first time.

Let us take the two cases that we have looked at before. There might not have been a conviction in the Stephen Lawrence case because the investigation was not done properly. That appears to be the implication of the report that followed the case. That might apply to the Damilola Taylor case, although that is less well evidenced by subsequent inquiry.

On the other hand, there are many cases in which the police have already started to produce scientific evidence that has been highly influential in bringing verdicts. I am a lawyer and my hon. Friend the Member for Somerton and Frome is a scientist, so I hope that he will shortly make a contribution on the science of the matter. I ask the Minister to put on record what the Government understand by scientific evidence if the amendments were to be accepted and the new evidence limited to scientific evidence. DNA is obviously the main type of evidence, although others could be included. What would the Minister exclude by way of new evidence, if the provisions were to be limited to scientific evidence?

I am most grateful to the hon. Gentleman—I can scarcely contain myself. I ask him to address two points. First, under the clause, evidence is new if it was not available at the time of the acquittal. Since DNA has been routinely available for

a number of years, would the hon. Gentleman's amendment not confine the operation of the provisions to older cases? Secondly, I ask him to consider the fact that his amendment would exclude the calling of a live witness who would be able to give new and compelling evidence.

I shall deal with both matters. First, there is an interesting debate about whether the DNA evidence was available in a case. I understand the hon. Lady's point, and I want to develop it. My understanding is that the discovery that allowed DNA to be used as evidence was made about 18 years ago. My further understanding is that it has developed into a significant element in investigation in that period. In addition to the original sources of DNA, such as skin or semen samples, the checking of other physical attributes, such as the nose, is being developed. The science is continually evolving.

It is perfectly proper to ask what the Government have in mind; whether new and compelling evidence—or whatever test is agreed to—is to be limited to scientific evidence only, and whether it would make a difference if that became new at the moment at which the investigation was done or was in theory available 18 years ago, although no one had done the work. There is now a database that holds some 1.5 million people's data. It holds the data of all those who have recent previous convictions. It can also hold data on people who have no recent previous convictions, or no convictions at all.

Last Session, the Government pushed through legislation, which we opposed, to allow data to be collected and held on people who are arrested and released. Evidence collected by the police as part of an investigation is no longer automatically destroyed when a person is released without charge or found not guilty. We opposed that legislation and continue to oppose it. It is entirely wrong that we should have a Big Brother state, which is not content to collect evidence on people who go through the criminal justice system, but collects evidence on people who are not found to have done anything wrong.

My first set of questions, therefore, relate to the question asked by the hon. Member for North Down. Would DNA come within the remit of new scientific evidence, and which other scientific evidence would come within that remit? In theory, DNA evidence could have been collected up to 18 years previously. Would such evidence still be considered new? Obviously, some inquiries will go back to acquittals that took place longer ago than that, if the people involved are still alive.

Would not that issue be covered by clause 66? Under that clause, the court must decide whether it is fair and in the interests of justice to use evidence from 18, 20 or 25 years ago. Such evidence may not be available to a court in a retrial because of the length of time that has elapsed.

We shall have to debate how fair it is to use evidence from a long time ago, but to be fair, that argument is not as strong as some others. If it is reasonable for inquiries to be made, 10, 20 or 30 years later, into whether Hanratty was guilty of the A6

murder, it is equally reasonable to reopen the case of someone who was acquitted that long ago. Obviously, the older the evidence is, the more necessary it is to regard it with suspicion, for two reasons. First, in the case of personal evidence, memories fade and confusion is more likely. Secondly, there is a greater risk that old scientific evidence will not be as pure, clear and unequivocal as new evidence.

The hon. Lady's second question was whether the amendment would exclude other non-scientific evidence, such as someone coming forward and saying, ''I haven't said anything before but I did it'' or—much more likely—''I saw someone do it but I did not say anything because I am their spouse and I felt loyal, but I cannot go on concealing the secret.'' I am probing to ascertain what sort of evidence the Government consider might be excluded, were we to limit this precondition to scientific evidence. Although there might be a strong argument for including evidence from a key witness who came forward later, we would not want to include such evidence if the only reason that that person did not come forward before was because the police never pursued that line of inquiry. It would be entirely wrong for the police later to produce a witness to whom they had not thought of speaking before. It would be different, however, if a witness to whom the police had spoken had said in evidence, ''The accused was in Brighton when the murder happened in Gateshead'', but later changed his story, confessing that his previous alibi evidence had been untrue. I understand the difference between the two scenarios. We must be sure that we do not countenance the inclusion of any evidence that could have been collected before but was not collected because the investigation was inadequate.

The third big question—although only partially relevant at this stage—is, how influential will new scientific evidence be? I do not claim to be an expert, but I have read about the probabilities of DNA evidence. Some papers that I read, and partially understood, talk through the theories of probabilities and Bayesian theories. They deal with the courts' decisions on when one can introduce arguments about probabilities into the scientific evidence debate. To put it bluntly, I am eager for the Minister to place on the record the Government's information on such matters.

Many people believe that DNA is unique to an individual. It is not; identical twins share the same DNA. Moreover, there is about a one in 3 million link in a population in this country of about 60 million. Therefore in every 3 million people there may be someone who has similar DNA. There are of course many arguments about how the jury, the prosecution or the defence might interpret DNA theories. Will the Minister give us a definitive statement on the rules concerning what courts may be told—and the crucial test for us is what the Court of Appeal would be told—about the relative and probative value of new scientific evidence, particularly new DNA evidence? That is fundamentally important.

I have two other points to raise at this stage. My understanding is that the scientific evidence may be of very good quality, but that damp or the conditions in which it is kept may damage its quality and probative

affect. Environmental circumstances such as where DNA evidence is collected, where it is held, and how long after the event it is collected may make it less influential. Dangerously, as a result of reading the popular and other press, the presumption is now generally abroad that DNA evidence, which was not used in a court case before, if produced will be conclusive. My understanding is that that is not the case. The public presume that the major new scientific evidence in this context is DNA evidence, but there are other sorts of new scientific evidence to be considered.

Those who look after the Forensic Science Service—I quote Dr. Bob Bramley, the custodian of the database—say of DNA:

''It is now recognised by the police as one of their most powerful intelligence tools.''

A spokesman from the Forensic Science Service added that the chance of finding two full DNA profiles that appear the same but which are not from the same person is possible but very slim. To return to the quotation:

''But we are very aware of this; it is not something we have not considered.''

There is all the difference in the world between probative or conclusive evidence, and strong evidence that contains the slim possibility that it may not relate to the defendant. It is helpful therefore to have a debate on opening such a large set of gates. The hon. Member for North Down knows that I, like her, need to be persuaded. I understand the strength of the argument about scientific evidence, but Parliament should not simply sign up and say, ''Yes, this is all fine. This will be the answer to all unsolved crime.'' We must investigate what scientific evidence it is that we are discussing. Therefore, I should be grateful if the Minister would deal with the concerns that we should limit any ''new'' evidence to that which is clearly persuasive, and certain scientific evidence falls into that category.

I wish that the Minister would also deal with the question raised by the hon. Lady as to whether DNA evidence would count as ''new'', given that the science has existed for around two decades, as well as the question whether it would be better, at least initially, to limit any retrial on new evidence to that sort of evidence. Moreover, I want the Minister to ensure that when we come to a judgment about those matters we can be satisfied about the strength of scientific evidence, so that proper assessments can be made of how effective the evidence might be, should be or would be in persuading the Court of Appeal that it falls into the categories that the hon. Member for Wrexham mentioned.

I am grateful for the hon. Gentleman's opening remarks. Undoubtedly, one of the clause's key issues is whether we should seek to limit to certain categories the new evidence that can be adduced.

I wish to provide some answers to the questions that the hon. Gentleman raised. I shall see whether the Minister agrees with them, because one feature of the

provision is that there is a belief in the mind of the public that scientific advances have led to a situation in which it is possible years after the event to come up with evidence that, once established, completely satisfies everyone that the person has committed the offence. By that I mean scientific evidence and, in particular, DNA evidence.

I shall seek to illustrate to the Committee that that belief is far from being a reality. DNA evidence, although it may be instrumental in securing a conviction, does not come up to that standard. It may in future, but it does not at present. However, before I deal with that, because that comes under scientific evidence, the Committee will see that I have lent my name to the amendments for scientific evidence alone to be admitted. Indeed, amendment No. 337 limits scientific evidence to evidence derived from DNA analysis. I shall come to that at the end because, on reflection, that is to go a step too far, for reasons that I shall explain.

To leave scientific evidence to one side, should we be limiting and excluding other evidence, which would effectively be the oral evidence and recollection of new witnesses? There is no doubt that, as the hon. Member for Southwark, North and Bermondsey said, the longer the elapse since the first trial, the more difficult it is to evaluate oral recollection.

First, it becomes extremely difficult to test that evidence by reference to other people or surrounding circumstances. Were I to tell the Committee that I had jumped up and down on the desk in front of me five minutes ago, it would not be difficult for hon. Members to conclude that I was not telling the truth. However, were I to tell the Committee that I had done the same thing on a visit to the Houses of Parliament 25 years ago, it would be extremely difficult to establish whether that comment was true or false.

Secondly, the longer the time that elapses, the more the recollection of an individual who is genuinely trying to help the court becomes a risky business to accept. There is nothing more misleading than the truthful witness who gives false or mistaken evidence—their truthfulness and integrity shine through because of what they say and how they speak, but it does not mean that they are right. For instance, I remember witnessing a road accident at the age of 18. Parts of it are engraved on my memory, but it would be a risky business to rely on that recollection today for details of what happened rather than the giving of evidence only three months later. Inevitably, because of the way in which it has located itself in my mind and remains there, it is likely to be affected by the passage of time, and important elements may have slipped out. Indeed, it is even possible, as one realises when one starts recollecting things from one's childhood, that others may disagree with one's recollection of certain events; one may have imagined that something had happened, yet genuinely believe that that was part of the reality.

We do not know whether retrials will be taking place one year or even two or three years after the previous trial, but the current state of scientific evidence suggests that the majority of the trials with which we are concerned are likely to concern episodes

that took place some time ago. I therefore have serious reservations about admitting new categories of evidence that, by their nature, are not capable of being analysed and tested objectively, as scientific evidence can be. The Committee must consider whether we should confine the evidence to scientific categories. I strongly suspect that few retrials will not depend on scientific categories.

My anxiety is that, having started with scientific evidence, prosecutors may be tempted to carry out trawls to see whether, many years afterwards, they can beef up a case with little titbits of oral evidence that might not previously have been available or that someone might have put forward. That could turn out to be a dangerous process. Of course, that may be precisely the sort of area that the Court of Appeal should deal with—although it is interesting to note that the Court of Appeal can choose what should be included or excluded at the subsequent trial. That is another issue on which I should be grateful to hear from the Minister.

Once the Court of Appeal has decided on a retrial based, for instance, on DNA evidence, is that the only extra bit of evidence that can be admitted? I am not by any means persuaded. We need some clarification. Having decided to admit extra DNA evidence, can the Court of Appeal say, ''There is a desire to admit the evidence of Mr. X, but we will not allow it''? That is something that the Committee needs to consider.

I now turn to the issue of scientific evidence. As is apparent from a brief study of the issue, DNA is by no means a foolproof method of carrying out identifications years after the event. The first reason was advanced very persuasively by experts who had been working for the Government, and who had advised the Forensic Science Service. That simple point that appears in the service's booklet, which states:

''The increasing sensitivity of DNA systems mean that individuals must be constantly aware of contamination issues. Items may be contaminated by handling without gloves or adequate precautions and from saliva transfer due to talking, coughing etc. over an unprotected item. Care must also be taken to ensure items do not cross contaminate by coming into contact with one another.''

In his opening remarks, the hon. Member for Southwark, North and Bermondsey mentioned the Hanratty case, in which the Court of Appeal decided not to overturn the conviction. However, it was apparent that although the scientific evidence appeared to suggest that the DNA on the key items connected with the murder was Hanratty's, the forensic scientists had a problem—as they maintained throughout the case, and as was explained to me in detail by a forensic scientist who was involved. The problem was that although the handling of the exhibits in the early 1960s was compatible with the standards to be expected at that time, the risk of cross-contamination was a live issue, which could not be excluded.

The fact that modern scientific handling methods were not used originally, because those concerned were not even looking for DNA, raised a real question in the mind of the scientist who spoke to me. He did not think that the Court of Appeal's refusal to overturn

the conviction was wrong, because there was other compelling evidence, but from the point of view of a scientist considering the matter in the abstract, he said that he could not possibly have said that he was satisfied as a result of the tests that Hanratty must have committed the crime. The handling issues, which it is impossible to resolve after 30 or 40 years—partly because of the lack of oral evidence about what was done with the samples—mean that a question will always remain. There are limits.

Cases of the type that may well arise to be dealt with under the clause will in many cases, I believe, involve the handling of material long enough ago for modern handling systems not to have been in use. Also, we cannot get away from the fact that if there is a challenge to the apparent DNA evidence on the basis that there might have been contamination, the passage of time will make it extremely difficult to deal with the matter in a way that is fair to the defendant. I do not believe that DNA evidence will prove the great crime-solver. I suspect that if the Court of Appeal is doing its job properly a number of cases may go before it with claims that DNA provides persuasive evidence, but that the Court of Appeal may feel constrained to say that a safe trial will not be possible.

The hon. Member for Southwark, North and Bermondsey raised the point that DNA evidence is a tool relating to probabilities, but historically it has been astonishingly open to misuse—so much so that the Court of Appeal issued a defining judgment in the case of Doheny and Adams in 1996. That is the most cautionary reading, which I recommend to the Committee as it is extremely comprehensive, about the abuse of DNA evidence by the prosecutor in presenting his case to the jury. More important, I think, there is a general public abuse of DNA evidence, relating to the understanding of what it can or cannot show.

The case established clearly that it is necessary to be very careful not to mix up the probability of an occurrence in the population with the probability that the person in question committed the offence. That is precisely what the prosecutor had begun to muddle. What emerges, as the hon. Gentleman rightly said, is that in truth a person's DNA is not unique, as far as the analysis that is currently possible goes, although in future it may become so. Perhaps it will become possible to refine it further. Therefore, it is likely that in most cases other people in the country—not necessarily identical twins—will have identical DNA to that of the person who is suspected of committing the offence.

The Court of Appeal laid down very strict guidelines, which are worth reading out.

''The scientist should adduce the evidence of the DNA comparisons between the crime stain and the defendant's sample together with his calculations of the random occurrence ratio.''

That is the occurrence within the population. In this country 25 or 26 people might have that occurrence. In addition:

''The Forensic Science Service should make available to a defence expert, if requested, the databases upon which the calculations have been based.''

I am picking out key quotes.

''The expert will . . . give the jury the random occurrence ratio—''

and nothing else.

''The expert should not be asked his opinion on the likelihood that it was the defendant who left the crime stain,''

nor should the prosecutor make suggestions to the jury that that can be concluded.

The issue is complex. The prosecutor argued that as only one person in a million has a DNA profile that matches that of the crime stain and the defendant has a DNA profile that matches the crime stain, ergo there is a million to one probability that the defendant left the crime stain and is guilty of the crime. That is a fallacy. The truth is that DNA can serve its purpose only by providing an indication that might help to exclude someone from the category of person who committed the offence. Otherwise, taken together with other evidence, it might be sufficient to satisfy a jury that a person is linked to the offence and has committed it. In many circumstances in which people have been occupying the same premises, particularly in cases such as rape, a defendant's DNA might be present quite innocently at the crime scene without contamination having taken place.

I am sorry to have taken so much time, but it is important for the Committee to understand that this is not a foolproof method by which, 20 years after the event, scientific knowledge suddenly enables one to turn round and say that an individual is guilty of an offence. It does no such thing.

Although I put my name to the amendment that requires scientific evidence to be limited to DNA, I did so merely for the purposes of probing. My view is that that is wrong. There are many other standards and categories of scientific evidence that might be of use to a court in reaching a reasoned decision, and those could develop further. Corneal evidence can be helpful in terms of identification, and I am told that ear prints might prove to be of interest. Voice recognition could develop to the point at which it is sufficient to give a clear indication of a person's identity.

We come to the old subject of fingerprints. They provide another great cautionary tale, which the Committee might bear in mind. When I was first at the Bar, one had only to find 16 matching features in a pair of fingerprints and it was considered to have been established with certainty that the fingerprint belonged to the individual. Now we are aware that that is not the case; it is much more complicated. The evidence can only give a degree of probability and point out how many matching features are present.

I am grateful to my hon. Friend the Member for Woking (Mr. Malins), who cannot, unfortunately, be here this morning. He attended the recent meeting of the Judicial Studies Board, which was shown an exercise in which some 40 scientists, all recognised as experts in their field, were given two fingerprint samples to compare. One expert saw 14 points of comparison and another 54, while the majority saw 20 to 25 such points. Public perception is that fingerprint

evidence is in some way foolproof. Although it has, historically, been viewed as the determining issue, it is nothing of the kind; it is merely a tool to aid understanding. I hope not to go over old ground, but the Minister will remember that earlier we debated the question of compelling defendants to hand over the names of the experts that they consulted, as a weapon, I think, with which they can be browbeaten when they go to court. I hope that my points can be a cautionary illustration of why nothing of the kind should occur, and that the Minister may even at this stage reconsider the matter.

Will the hon. Gentleman address again the issue of clause 66? We cannot consider clause 65 in isolation. The new and compelling evidence that must be considered must be weighed against the needs of a fair trial. Clause 66(2)(b) mentions

''the length of time since the qualifying offence was allegedly committed''.

The hon. Gentleman referred to older cases, such as those from 20 years ago. However, that is ruled out by clause 66.

The hon. Gentleman also referred to DNA and fingerprint testing. Does he think that the court will again rule that there will not be a fair trial under clause 66(2)(c) if, as specified, that evidence was available at the time,

''but for a failure by an officer''

to take it? Fingerprints, DNA and time rule out a great swathe of cases. Which cases does he think those clauses will apply to?

The hon. Lady makes a good point. An argument that was put against the procedure that we are embarking on a process to change old systems is that the number of cases covered is likely to be extremely limited. I have always thought that, and the hon. Lady has heard my views. We are being responsive to public pressure—perhaps excessively so—to see that justice be done in cases in which it appears manifest that someone is guilty of an offence of which they have been acquitted.

The passages that the hon. Lady quoted from clause 66 are good safeguards, and I am pleased that they are there. I would assume that in many cases people will be quite surprised to discover that cases cannot go beyond the Court of Appeal—they might not even get to the Court of Appeal if the Director of Public Prosecutions does his job properly. People might simply conclude that there is no possibility of a fair trial. That is a very real issue. There is a belief among the public—which has been fuelled by politicians, I am afraid—that the provisions will provide a great new way of solving injustice and doing justice to the victims. However, I suspect that the provisions will be limited in scope. That said, however, there might be cases in which, notwithstanding all the matters that the hon. Lady described, the tests can be satisfied. Speaking personally, I shall not stand in the way of the Government as long as the tests are satisfied and the legislation is satisfactorily limited.

The hon. Gentleman referred to the fact that there has been some political fuelling of the

belief that the provisions will suddenly produce a load of convictions of people who have previously been acquitted. Does he share my concern that senior police officers also appear to have been too enthusiastic about the prospect of dealing with a lot of unsolved crimes? Some have strayed into incautious remarks, suggesting that the provisions will manage to bring justice in cases in which justice has not previously been satisfactory. Does the hon. Gentleman think that they ought to be more careful?

Senior police officers are like any other member of the public: they have their emotions and feelings. I cannot believe that the average chief officer who makes such proclamations has been to the depot—wherever it is—where the Government scientists keep the samples and exhibits, or that such evidence has not been examined or is unlikely to be examined again. I suspect that the remarks that the hon. Gentleman referred to are expressions of hope. However, I agree that the trouble is that people's hopes can be raised well beyond what is possible.

Is not this series of amendments, which would limit new evidence to scientific evidence, ill advised? I ask that particularly in the light of the example that the hon. Gentleman has just given of 40 different fingerprint experts with 40 different views. Would their views constitute scientific evidence? I rather doubt it. Their results may have been detected with a magnifying glass and a tin of powder, but that does not make their views scientific—they are purely a matter of opinion. Some advance in psychiatry might produce psychiatric evidence that was not available at the outset, but psychiatry is described very much as an art, not a science. The hon. Gentleman is therefore in danger of excluding a great deal.

The hon. and learned Lady makes a good point. I would consider fingerprint evidence to be scientific evidence, although all scientific evidence is ultimately a matter of opinion. A series of analyses is carried out on material or an occurrence, and someone gives an opinion as to the conclusions that can be derived, so there is inevitably a human dimension. As I said, however, the merit of limiting evidence to scientific evidence is that it would exclude the possibility of simple oral recollections of facts and events being brought in. The proposed wording could perhaps be different, but we are trying to concentrate minds on the key issues.

The hon. and learned Lady makes a perfectly reasonable point, and I shall be interested to hear how Committee members view the safeguards, and how they see the process working in practice. I shall also be grateful to hear from the Minister.

I have taken up quite enough of the Committee's time, although I hope that my points will be of use. This is a difficult issue. I am convinced that there will be far fewer retrials than the public think. I am also convinced that scientific evidence is not without its difficulties, from which I conclude that I would not want to limit scientific evidence to DNA. All scientific evidence—or forensic evidence, depending on what one wants to call it—has shortcomings, but one could argue that evidence based on human recollection has even more.

For once, lawyers do not have all the answers. Some of our learned colleagues are in great danger of mixing up science and technique, which are quite different concepts. If we are not careful, we shall get into yet another debate about words, which will lead absolutely nowhere.

What matters is the quality of the evidence—whatever form it takes—and whether it satisfies all the other tests in clauses 65 and 66. If those tests are consistently and thoroughly applied, it will not matter whether we are talking about an object, forensic evidence or even oral evidence.

If we are going to describe the evidence required, I would prefer us to use the word ''forensic'' instead of the word ''scientific'', because it is much more specific. The science is in assessing the quality of the evidence; it relates to the method applied, not to the technical details. It is the quality of evidence that matters. I therefore strongly urge the Minister to resist the amendments, which would create far more difficulties than they resolved.

My hon. Friend the Member for Southwark, North and Bermondsey kindly described me as a scientist, but I would not describe myself as anything of the sort. My first degree was in a scientific subject—physiology—but that perhaps only qualifies me to lay claim to an understanding of some of the principles and vocabulary involved.

We are seeking to probe three things with the amendments. First, what sort of evidence, outside the field of scientific investigation, would the Minister consider suitable to be brought forward? Clearly, it would have to be relevant to the context of the original acquittal. The hon. Member for North Down has already drawn the Committee's attention to clause 66, which is relevant to this question. There is a time-expired quality to much of the evidence that is not the result of improved techniques of one sort or another; that would, I think, limit the amount of other evidence that might be adduced and might pass the test in clause 66. It would be helpful to hear from the Minister what evidence he would see as relevant.

When the review of double jeopardy was instigated, the original argument for re-examining a long-established principle was that we now had the whizzo new technique of DNA testing and it was wrong not to use it in court proceedings where it could make a difference to the quality of the evidence. However, the Bill will extend the revision much wider than the use of DNA evidence, so we need to know what the Minister has in mind.

One of our amendments would limit the definition of scientific evidence to DNA testing; I want to dwell for a moment on such testing, which is an extraordinarily powerful method of providing evidence. I think that everyone accepts that it is a valuable tool, at least in investigative terms, and that it has some value as evidence, in securing convictions. However, difficulty arises in conveying to a jury, and even to the odd judge, the value of that evidence and the fact that it is not incontrovertible.

There is not one sole explanation to be drawn from DNA matching. These matters bring the argument into the realm of probabilities and Bayes' theorem, which my hon. Friend the Member for Southwark, North and Bermondsey referred to. It relates to the assessment of the basic probability of a match, but also has a role in relating that to the context and circumstances to which probability is being applied. There are difficulties, however, and the legal system has introduced its own to the process.

In referring to the case law, the hon. Member for Beaconsfield might have quoted the conclusion of Lord Bingham of Cornhill in 1998 in R. v. Adams, when he was asked about the admissibility of evidence on the statistical application of Bayes' theorem and the mathematical formulae—they are very complicated, and beyond me as a mathematician. He said that the introduction of mathematical formulae was

''a recipe for confusion, misunderstanding and misjudgment, possibly among counsel, probably among judges and certainly among jurors''.

Let us set aside for a moment the assessment of the relative numeracy of those different classes of individual. I suspect that many jurors would prove more numerate than those trained in the legal profession, who do not normally have a scientific or mathematical background, although that is just a cultural phenomenon. The conclusion reached by Lord Bingham means that many relevant cases are based on the assertion of probability, rather than on an understanding of the context in which the probability was deduced, enabling the jury to reach its own conclusions.

Under current techniques, there is the possibility of more than one person being identified under the same DNA analysis. That may be so for the world's population, but it is a basic fallacy that we should deal with the world population. Considering the probability of a match among the world population is nonsense, because DNA types are more likely to be similar in a closely related population. Taken to the extreme, in a population that is interrelated because of lack of movement over a long time, it is much more likely—there is a much higher probability—that similar DNA patterns will emerge.

For example, in my area of the country, the Somerset levels—I can say this, because my family originates from there—people are remarkably interbred. That has medical manifestations. I used to practise in Burnham-on-Sea, and I could tell which of my patients came from the levels because they had ophthalmic abnormalities. Those congenital abnormalities were the result of the gene pool being considerably smaller in that area. Each community was originally on an island and people did not move from theirs, or if they did, they moved only to the next island.

I offer the following illustration. A little while ago, the remains of a cave dweller in Wookey Hole were analysed for DNA purposes and, as an experiment, an analysis was carried out on some local people to see whether there were any similarities. Lo and behold,

Mr. Targett, a history teacher at the Kings of Wessex school in Cheddar, was found to have such a close DNA match to the cave dweller, who lived many thousands of years ago, that it was thought highly probable that his family had, over about 4,000 years, achieved a total migration of about 4 miles. I do not ascribe that to lack of ambition and I relate the story only because it illustrates the fact that in a small community there is a much higher risk of mismatching due to the much higher probability of similar DNA traces.

DNA evidence is normally used to corroborate identity or location, and to argue against alibi evidence. Inherent in that is the possibility not only of an innocent explanation—perhaps it is the result of contamination; perhaps the genetic material belonging to the individual comes to be there quite properly—but of the DNA belonging to a different individual who happens to have a close genetic match to the defendant. It is important to recognise that.

I move on to the other scientific evidence that might be adduced. As the hon. Member for Beaconsfield said, it is a burgeoning field. Biometrics will play a much bigger role, not only in the investigative and judicial processes, but in matters such as security, and a lot of work has been done on it here and in America. On Friday, I visited the university of Kent at Canterbury, which now has a forensic science department. That is a good thing, because it is important that we keep up to date with developing technologies.

The Police Information Technology Organisation has commissioned a lot of that work, and although there are many definitions of biometrics, the organisation's definition describes it in this way:

''The automated identification or verification of human identity through measurable physiological and behavioural traits.''

Consider the second part of that—behavioural traits, which bear on the point made by the hon. and learned Member for Redcar. I have serious problems with the state of psychology if it accepts them as incontrovertible evidence of anything. Psychology is a young, developing science, yet the door is slightly ajar for developing psychometric analysis techniques to be adduced as compelling evidence.

The experience of psychological researchers is that for every genuine advance there is a cul-de-sac. The cul-de-sac route might lead to appalling results for a defendant if it were adduced as compelling evidence. Even hard science, which considers physiological traits, includes iris corneal recognition patterns, ear prints, handwriting analysis and voice recognition techniques. The problem with some of those is that they can be simulated to avoid detection, and for every bit of technological gadgetry that is used to recognise voices there is the equal possibility of producing a gadget that will disguise them.

The basic problem is how to assess the evidential value of what appears to be a foolproof scientific technique. Most people who study the processes will say that science is never foolproof, that it is always a matter of probabilities and that no one can be certain about anything. All that can be said is that a theory

has been tested and so far has been proved correct, so the probability is that it will be correct the next time also.

Is photographic enhancement—the use of computers and image recognition vastly to improve the recognisability of an image on a photograph or on film or video footage—an acceptable scientific advance? That brings me back to the matter raised by the hon. Member for North Down. Is the fact that a film can now be better analysed than when it was available to police at the time of the original trial new evidence? I suspect that it is. If so, at what point do we determine that that new technique can be interpreted as new evidence? That is a critical factor.

We might also consider techniques other than those used in identity recognition—better computer analysis of financial data, for example. One problem with serious fraud cases is that there may be a vast amount of information to be analysed. Perhaps it may be analysed at an early stage, but only if the entire resources of the country were put to work. However, at a later stage, there might be a single computer that could do the same job in days. Is evidence new if it is arrived at by a technique that is not, in essence, different from that used originally, but which has simply been speeded up by the application of new technologies?

The hon. Gentleman is making some interesting points, but the definition of new evidence is set out in clause 65(2):

''Evidence is new if it was not available or known to an officer or prosecutor at or before the time of the acquittal.''

That is precisely the point. The example that I cited involved analysing large amounts of material. The evidence was available, but it could not reasonably have been obtained because the resource implications would have been too substantial. Modern techniques would make it possible to adduce that evidence by running all the receipts through a powerful computer capable of recognising patterns that an individual could recognise if his brain were big enough. Certain questions of definition need to be considered. I certainly do not believe that we should be too prescriptive, but perhaps the Minister will expand his world view to all the possibilities and give thought to how we could provide better definitions.

I end by making a suggestion. Because technology in this area is developing so fast and because it will not be clear to people engaged in the law—in whatever capacity, including jurors—how to assess the validity of evidence, the subject should be kept under review. However, I do not know whether the Law Commission could set up a technology sub-committee to review new techniques and give guidance on their application in forensic work. I take issue with the hon. Member for Brighton, Kemptown (Dr. Turner), because ''forensic'' simply means relating to a court, so a definition including the word ''forensic'' would have no meaning at all. That word is often used as shorthand for science, but in the phrase

''forensic science'', the word ''science'' means science and ''forensic'' refers to court procedure.

There is a case for such a review, which could give clear guidance to judges on the evidential and probative value of certain techniques and regularly update them. Without that, I fear we shall see yet another example of the legal system finding it difficult to keep pace with the rate of change in science and technology, which, I believe, will gallop ahead in the next few years.

I urge the Minister to resist the amendment that would limit the evidence to the scientific. Our debate seems to have moved from that question to wondering whether science can ever produce sufficiently compelling evidence, even if we limit the evidence to the scientific, although that is somewhat difficult to follow.

The hon. Member for Somerton and Frome suggested that new evidence should pass tests set by the Law Commission before being admissible, but new scientific evidence usually has to fulfil certain criteria before being accepted by the courts, so that it does not become a major problem. We are discussing many issues that the Court of Appeal will have a duty to consider in deciding whether evidence is compelling. All the arguments made on the amendment's weaknesses can, and no doubt will, be advocated by representatives of legally aided defendants.

Scientific advance has been important in building up the momentum behind this change. Developments in the use of DNA evidence have made us all appreciate how an absolute bar on retrial can be extremely unjust to victims. I had intended to take the hon. Member for Southwark, North and Bermondsey to task for talking about the Big Brother state and for unfairly criticising the police and Crown Prosecution Service over the abortive Stephen Lawrence prosecution. That was a private prosecution, brought against the advice of the police and the CPS. It was brought by my head of chambers; it was also brought against my advice—just to show how much notice anyone takes of me. It is not right to blame the state for that.

The Committee will recollect that last week I referred to a campaign waged by the mother of a murder victim. However, if scientific evidence were the only way to trigger a fresh trial, her case would fall outside that. I spoke to Ann Ming, the mother of Julie Hogg, who was murdered by the man who was acquitted, on the telephone yesterday. She said—I do not pretend to understand the terms in which this occurred—that that individual gave evidence that he had killed her daughter. If that is not a compelling case for a retrial, I do not know what is. It has nothing to do with science.

The hon. Member for Somerton and Frome invited me to expand my world view. Let me tell hon. Members that there is not a nanosecond that I spend in Committee during which my world view is not expanded. That has certainly been the case in this debate, which I can only describe as educational. I have learned a great deal about the genetic make-up of the hon. Gentleman's constituents in Somerset, which

I had the great pleasure of visiting as a member of the Environment, Transport and Regional Affairs Committee.

I have certainly learned a great deal about DNA from the debate, but my hon. and learned Friend the Member for Redcar hit the nail on the head when she said that these are issues that the court system and trials have to grapple with all the time. The point is that many issues raised in the debate will be matters for a retrial should the Court of Appeal order one. In that sense, they do not raise additional issues concerning double jeopardy.

The test for the Court of Appeal—my hon. Friend the Member for Brighton, Kemptown was absolutely right to talk about the quality of evidence—is that it must be new. I accept the example that the hon. Member for Somerton and Frome gave of evidence that was there, but which could not be detected as the technology did not exist at the time.

DNA evidence is a good example of that. Bloodstains and other genetic samples may be available, but we would be unable to tell to whom the blood or genetic sample belonged until the appropriate technology was developed. The Court of Appeal will have to judge whether evidence is new and compelling, and that is defined in clause 65(3)(a) as ''reliable'' and in clause 65(3)(b) as ''substantial''. In layperson's terms, evidence must show a high probability that the person did it.

The hon. Member for Southwark, North and Bermondsey asked about the rules governing the use of DNA evidence. He got the answer from the hon. Member for Beaconsfield, who read out information on a case in the Court of Appeal.

Yes, Doheny and Adams.

Indeed. That guidance is offered to courts when they are considering such evidence.

The hon. Member for Southwark, North and Bermondsey asked what the Government understand by ''scientific''. The honest answer is that I do not know. That is precisely why the Government have not offered the word ''scientific'' in the clause; it is also why the amendment is defective. I do not think that it is possible, and the debate has highlighted it, to define ''scientific'' as a category separate from other evidence. New, compelling evidence—scientific or other—that shows that someone has ''got away with it'' should be considered.

The hon. Gentleman, and the hon. Member for North Down in her intervention, answered the question as to what would be excluded. One thinks of certain circumstances—finding a murder weapon after many years, for example. It is also possible for a witness to come forward after many years to say, ''I was married to him at the time. He came back that night covered in blood and said to me, 'I've done her in'. I have kept my counsel for 20 years, but I have decided that I can no longer live with myself and have come forward.'' If such evidence were available, it would be wrong not to consider, first, whether it meets

the new and compelling test and, secondly, if the Court of Appeal is so satisfied, whether it should be included to be considered alongside all the other evidence in the retrial. That is why I do not accept the amendment.

The point about protection against sloppy investigating, which the hon. Member for Southwark, North and Bermondsey made, is covered by the due diligence test at clause 66(2)(c).

Finally, the hon. Member for Beaconsfield asked whether the Court of Appeal will be able to decide that new evidence A can go in, but not new evidence B. That is not a decision for the Court of Appeal. It must decide whether new and compelling evidence exists. All the evidence that the prosecution wishes to present at the subsequent retrial will be included at that retrial, and the court and the jury will consider it in the normal way.

The Minister is right in his analysis that, after the amendments were tabled, I became increasingly conscious of the fact that they present difficulties. Nevertheless, they are a legitimate area for discussion. They may highlight the fact that this is an all-or-nothing situation, and that one is either prepared to accept the principle of retrial that the Government seek or that one should stand up against that principle. I accept that the difficulties of compartmentalisation, with which the amendments deal, would be tricky to resolve.

The Minister's final comment, which confirmed what I already thought, gives rise to some anxieties. The public perception—indeed, perhaps the Committee's perception—of the principle of retrial is that the Court of Appeal has found something that justifies the case beginning again. That may be the justification, but thereafter the prosecution, within the ordinary rules, may seek to present other evidence. It would be the trial judge's decision as to whether that evidence should be excluded or admitted. On the basis of oral evidence and recollection, it would be extremely difficult for the trial judge to exclude it. Surely it is the case—I am widening the area under discussion slightly, but the Minister raised the point—that in ordinary circumstances a long delay in reviving oral evidence would be subject to an abuse of process application, which would prevent a trial from taking place. Such an application would be made to the trial judge.

However, it is not suggested that such an application can be made in these circumstances. Once the Court of Appeal has decided that the evidence presented is so compelling that a retrial should take place, I assume that that discretion, which would be the ordinary protection that might exist, would no longer be available. It would be difficult for a judge to exclude oral evidence. He would have to leave the matter to his directions and to comments from the defence on its reliability. That is an interesting area that the Minister may wish to consider. As so often in the Committee, one discussion triggers another.

I shall not press the amendment to a vote, as I wish to think about the issue further. I accept that the compartmentalisation of scientific evidence from other

evidence presents as many problems as it seeks to solve.

I am grateful for the debate, which leaves on the record the fact that scientific evidence itself will not be conclusive. If it could be defined, it is evidence that must be taken into account with all the other evidence. If we have at least managed to help to educate those who have taken the view that DNA evidence is the answer to a huge number of unsolved crimes, our debate has been helpful. It is a useful reminder that we shall have to work to define those offences that could be retried. We had a debate on that the other day. Furthermore, we must define the criteria for that, and that work began in this morning's debate.

That is a useful conclusion. We have yet to deal with other matters that deal with double jeopardy. In the light of that, I beg to ask leave to withdraw the amendment.

Amendment, by leave, withdrawn.

I beg to move amendment No. 264, in

clause 65, page 40, line 7, leave out from 'new' to end of line 8 and insert

'if its availability was not, and could not possibly have been known to an officer or prosecutor at or before the time of the acquittal.'.

With this it will be convenient to discuss the following amendments:

No. 331, in

clause 65, page 40, line 7, leave out 'or known'.

No. 340, in

clause 80, page 49, leave out line 27.

Amendment No. 263 raises a small point that merits consideration, although it is partly answered by clause 66. Clause 65 states:

''The requirements of this section are met if there is new and compelling evidence that the acquitted person is guilty of the qualifying offence . . . Evidence is new if it was not available or known to an officer or prosecutor at or before the time of the acquittal.''

The amendment would change the wording to

''could not possibly have been known to an officer or prosecutor at or before the time of the acquittal.''

That would reinforce the original due diligence test that applied to the officer and the prosecutor. It would not allow incompetence to be remedied by subsequent proceedings if evidence could readily have been brought before the court.

I am mindful of the fact that the question to be determined under clause 66(2)(c) is

''whether it is likely that the new evidence would have been available sooner (either in the earlier proceedings against the acquitted person or subsequently) but for a failure by an officer or by a prosecutor to act with due diligence''.

However, that is only a factor to be taken into account, whereas amendment No. 264 would effectively place a bar on reopening proceedings if it could be shown that the officer could have known about the evidence. That is the basis for the amendment.

I want to make two points. First, I support the hon. Gentleman's amendment. For the reasons that have been set out, we would do well to tighten the definition, and that is the purpose of amendment No. 331. The clause is drafted in such a way that it would include evidence that was available but unknown to the officer or the prosecutor. It opens the door to incompetence and a lack of due diligence—I can see no other interpretation.

Secondly, the test must be whether the evidence was there, not whether someone found it or failed to go down the relevant route of inquiry. However we resolve the issue of double jeopardy, I would have considerable concerns if trials could be restarted just because the prosecutor, the Director of Public Prosecutions, the CPS or the police happened not to know of the evidence, even though it was available. The issue must be the availability of the evidence, not the knowledge of the individual officer or prosecutor.

In essence, the debate is about the phrase ''could not possibly'' versus the phrase ''was not actually available''. As the hon. Member for Beaconsfield said, amendment No. 264 would restrict new evidence to that which was not available and could not possibly have been known to the police or prosecutors at or before the time of acquittal. That would be too restrictive, because evidence may subsequently come to light that could have been unearthed before the original trial, although, for whatever reason, it was not. For example, as we discussed, a witness may come forward who had not done so previously. Under the amendment, one could argue that such evidence should be excluded from consideration if it could ''possibly'' have been known to the police. I do not agree, however, and such evidence should be available to the court. As the hon. Gentleman said, the combination of clause 65 and the additional safeguards in clause 66—they are important, for the reasons given by the hon. Member for Southwark, North and Bermondsey—means that the court would have to take account of due diligence by the police and the prosecution in the investigation.

Similarly, I am not persuaded by amendment No. 331. The police might well be aware of a piece of evidence without being able to obtain it. A witness might, seemingly, have disappeared, or the police might have been unable to produce him. If the witness were to come forward subsequently, offering compelling new evidence, the DPP should be able to offer that as new evidence in the application for a retrial. In that instance the same protection would apply, provided for by clause 66(3)(c).

I understand the argument, but there would be all the difference in the world between the fact that an officer knew that there might be evidence but was unable to track it down or persuade someone to talk, and the fact that evidence existed but the officer or prosecutor never inquired about it. I should be grateful if the Minister could study the wording. I understand that clause 66 contains due diligence protection, but it still seems to me that we need to ensure that we shall not be making use of evidence that was thought to be out there, but simply was not in the hands of the prosecuting authorities.

That would open wide the range and number of cases that could be dealt with under the Bill.

The hon. Gentleman makes a vital point, because only the Court of Appeal will be able to make a determination under clause 66 and apply the due diligence test. Rulings by the DPP would be under clause 65 alone. A due diligence test should be included in that clause too.

The hon. Lady has made a point for me that I hope the Minister will attend to.

May I add something that has a direct link to that? We need a threshold that works in relation to the principle as well as rules that guide the Court of Appeal. We should be clear about that. Those are not to be confused. I understand that that is why the original drafting separated them. It is not a question of difference in principle, but the hon. Lady has made the point that real clarity is needed about what rules would allow new evidence to count and what thresholds, or other considerations of justice, would allow the court to decide that the case could pass beyond it and be reconsidered. There are two proper considerations and the public and Parliament would be well served if we kept those separate.

The Minister may want to respond now or to reflect and respond when we reach the end of our deliberations on the clause.

I fully accept the Minister's comment that the amendment might impose an excessive restriction. However, an issue that I do not find easy arises in this context. Perhaps the hon. Member for Southwark, North and Bermondsey has a better approach to the matter. Clearly a trigger is necessary, and that is provided for in clause 65.

Without that, it would not be possible even to start the process. There is a slightly curious aspect of this matter. I rather tend to the view of the hon. Member for Southwark, North and Bermondsey that the provision stating that

''evidence is new if it was not available'',

is straightforward, and clause 66 enables the reasons for the lack of availability to be examined. However, the insertion of the words ''or known'' in clause 65(2) seems to open an enormous gateway. Although the provision is qualified in clause 66, it appears, to an extent at least, to override it—to allow for an officer to say, ''I did not know, so I am entitled to make the application; the court can decide whether my reasons for not knowing are good enough to allow the retrial.'' I think that there may be an argument for removing the words ''or known'' as amendment No. 331 would do. Doubtless the Minister can go and reflect on the matter, as I hope he will. If he has not reflected on it by Report, we may tackle it again.

I beg to ask leave to withdraw the amendment.

Amendment, by leave, withdrawn.

I beg to move amendment No. 308, in

clause 65, page 40, line 17, leave out subsection (5).

With this it will be convenient to take amendment No. 379, in

clause 65, page 40, line 18, after 'have', insert 'not'.

When I first read subsection (5) I did not find it easy to understand.

It being twenty-five minutes past Eleven o'clock, The Chairman adjourned the Committee without Question put, pursuant to the Standing Order.

Adjourned till this day at half-past Two o'clock.