[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
	The Law Commission
You are here: BAILII >> Databases >> The Law Commission >> Double Jeopardy and Prosecution Appeals Part IV [2001] EWLC 267(4) (15 March 2001) URL: http://www.bailii.org/ew/other/EWLC/2001/267(4).html Cite as: [2001] EWLC 267(4)

[New search] [Help]

PART IV
NEW EVIDENCE AND THE DOUBLE JEOPARDY RULE

4.1      In Part III we explained that Article 4(2) of Protocol 7 to the ECHR, by way of exception to the rule against double jeopardy, permits the reopening of an acquittal where new evidence of the defendant's guilt has become available. In this part we consider whether, and if so in what circumstances, such an exception should be introduced into English law.

Should there be an exception for new evidence?

4.2      The crucial question is whether the principles underpinning the rule against double jeopardy can ever be outweighed by the need to pursue and convict the guilty. In favour of an exception, we can identify a high value in terms of the accuracy of the outcome of the proceedings – that is, convicting the guilty, and only the guilty – which is a key aim of the criminal justice system. To justify an exception, the advantages in terms of accuracy of outcome must override the collective and individual process values served by the rule.
4.3      In CP 156 we identified four process values: the risk of wrongful conviction, the distress of the trial process, the need for finality, and the need to encourage efficient investigation. We postulated some cases where the argument for allowing a retrial on the basis of new evidence seemed strongest, and concluded that the justifications for the rule were not such as to require an absolute ban on reopening an acquittal even in such a case.^[1] We therefore provisionally proposed that the double jeopardy rule should be subject to an exception in certain cases where new evidence is discovered after an acquittal.^[2] We then went on to examine individual features of our hypothetical strongest case to assist us in formulating the limits of the new exception we proposed.^[3]

The case for an exception

4.4      CP 156 put the positive case for a double jeopardy exception largely in terms of accuracy of outcome. Professor Ian Dennis has recently taken up this theme and has linked it to another important issue, namely the erosion of the legitimacy of the initial acquittal in the light of new compelling evidence of guilt:

We are … used to the idea that new evidence of innocence, a previously unknown alibi witness for example, calls into question the legitimacy of a conviction. It suggests that a mistake has been made that calls for investigation and possible rectification. Similarly the emergence of significant new evidence of guilt calls into question the legitimacy of an acquittal. It suggests likewise that a mistake has been made. Why should we not investigate and if necessary rectify the mistake, so as to lead to a retrial? It is not apparent now that a different verdict on a second trial would be inconsistent, given significant new evidence. The criminal justice system exists to enforce the criminal law, and the correct enforcement of the criminal law against those whom we have reason to believe may be guilty is a matter of state policy. The interests of justice seem therefore to call for a retrial in these circumstances. A retrial will resolve the legitimacy problem of the first acquittal and forward the aims of criminal justice if the defendant is in fact guilty.^[4]

4.5      There is, further, the spectre of public disquiet, even revulsion, when someone is acquitted of the most serious of crimes and new material (such as that person's own admission) points strongly or conclusively to guilt. Such cases may undermine public confidence in the criminal justice system as much as manifestly wrongful convictions. The erosion of that confidence, caused by the demonstrable failure of the system to deliver accurate outcomes in very serious cases, is at least as important as the failure itself.

The response on consultation

4.6      The issue whether an exception could be justified, and our conclusion that it could, attracted much robust discussion in the responses to CP 156. In total, 51 responses supported our proposal and 32 opposed it. This is indicative of the fact that the issue is both complex and sensitive. As a snapshot, the judiciary was split on the question, with divergent views being expressed by judges of the highest rank. A clear majority of individual judges were in favour of an exception, but the Council of HM Circuit Judges was against. A clear majority of individual practitioners were in favour, as well as two circuits of the Bar. The Criminal Bar Association, the London Criminal Courts Solicitors' Association, the Criminal Law Committee of the Law Society and the Law Society of Scotland were against. Police and prosecuting bodies were overwhelmingly in favour. Pressure groups and academics were mostly against. Members of the general public were equally divided.
4.7      The reasons given by those who supported our proposal were largely the same as those that we gave in CP 156 – essentially, that (in some cases at least) accuracy of outcome is more important than finality. That is not a point which needs, or is capable of, much elaboration. We merely note at this point that a clear majority of respondents shared our judgment on it.
4.8      On the other hand we attach great importance to the views of the substantial minority who disagreed with that judgment. All of these responses were cogently argued and required us to look again at our reasoning. We now examine the objections raised, and consider whether our provisional proposal can be defended against them, or modified in such a way as to meet them.

The case against an exception

4.9      Those opposed to reopening acquittals frequently cited, as the reasons for their opposition, the justifications for the double jeopardy rule which we had set out in CP 156. Consistent themes were the principle of finality of criminal proceedings, undue distress to defendants and the acquitted, difficulty in ensuring a fair second trial, the risk of wrongful conviction, the need to encourage probity and diligence in the investigation of crime and the need to ensure that the prosecution presents the whole case. One academic respondent predicted public anger at the reopening of acquittals.^[5]
4.10      What was most apparent from the arguments put by those opposed to a new exception was that we did not give sufficient weight to the importance of finality as a value in the criminal law. In CP 156 we largely identified the value of finality as a species of our concern with the distress and anxiety suffered by defendants.^[6] In this sense, finality was necessary to allow the acquitted defendant "repose" – freedom from the fear of having to go through the ordeal of trial again at some time in the future. We did not accord substantial value to any independent notion of finality. We are now persuaded that we were mistaken in this respect.

Finality as antidote to distress and anxiety

4.11      A very senior judge argued that it was "important to preserve the principle that a defendant acquitted by a jury need not worry that he may have to undergo the trial process all over again." In CP 156 we had argued that it would only be the guilty who would fear the future reopening of an acquittal, though we accepted that there was a danger that others would be subject to an ill-founded fear of reopening.^[7] We now accept that this latter danger may be greater than we recognised. We stated that "the reopening of a prosecution would by its nature be a very occasional occurrence".^[8] This may well be the case but, unless the limits of the reach of any exception to the double jeopardy rule were clear cut and notorious, the potential lack of finality and the associated distress and anxiety would affect a much larger group of acquitted defendants than would ever be proceeded against under the exception. This larger group would itself be composed of two groups: those to whom the exception could technically apply, but in respect of whom it would never in practice be invoked, and those to whom it did not even technically apply, but who might nevertheless fall prey to anxiety that it might. If, as we provisionally proposed, the exception were to apply to all cases in which the sentence would be likely to be at least three years' imprisonment, the category of defendants to whom it would be capable of applying would be very large and ill defined. Even for the second group, to whom the exception would not apply at all, we may have been wrong to describe their fear as necessarily "ill-founded". Even were the initial exception narrowly, specifically and notoriously drawn, it might be added to. By way of example, the history of extensions to the power of the Attorney-General to refer unduly lenient sentences to the Court of Appeal suggests that government might wish to extend the range of offences to which the exception applied.

Finality and individual liberty

4.12      We are now persuaded that there is an important sense in which finality as a value can impact on individual liberty or autonomy. In a liberal democracy, it is a fundamental political and social objective to allow individuals as much personal autonomy as possible, to allow people the space to live their own lives and pursue their own visions of the good life. Lack of finality in criminal proceedings impinges on this to a significant degree, in that the individual, though acquitted of a crime, is not free thereafter to plan his or her life, enter into engagements with others and so on, if required constantly to have in mind the danger of being once more subject to a criminal prosecution for the same alleged crime. We did not recognise in CP 156 that autonomy or liberty in this sense is to be valued for its own sake. We only recognised want of finality as a cause of distress and anxiety. Reducing the personal autonomy of the individual may, of course, occasion distress and anxiety, but that is not the only reason for valuing it.
4.13      The importance of finality in civil law is derived from the importance of having reasonably settled property and other private law rights. Such rights themselves are justifiable, in part, as important for personal autonomy. Just as the independent value of finality in the civil law, which we recognised in CP 156, has its roots in personal autonomy, so too is it so in the case of the criminal law, in which the relationship between a final freedom from pursuit by the state on a particular matter and personal autonomy is more direct.

Harassment by state officials

4.14      The value of double jeopardy protection is underlined by a further, related but distinct, argument, put by a number of respondents, which did not feature at all in CP 156. If a power to reopen acquittals existed, it could be used illegitimately by ill-intentioned state servants. Peter Mirfield wrote that the present rule "denies the prosecution the opportunity for oppressing the acquitted person simply because it believes the acquittal to be unjustified." Anthony Edwards, a very experienced and highly respected criminal solicitor, put it in this way:

Any but the most limited changes in the law will be used by some investigators to intimidate persons whom they believe but cannot prove to have committed crime. This may tend to concentrate on those from disadvantaged and minority groups.

4.15      Liberty pointed up a further potential disparity with regard to dealings with prosecuting authorities:

If the prosecution promised a defendant not to prosecute, proceedings would be stayed as an abuse of the court's process, whereas if a jury acquitted the defendant there would be no bar to a further prosecution.

Finality and third parties

4.16      We recognise that there is some value in protecting certain third party interests by finality of criminal proceedings. Some such interests, for example the reliance placed on an acquittal by an employer, may be of only marginal importance. More weight may properly be given, however, to the emotional and financial interests of an acquitted person's family and dependants.

Finality as wider social value

4.17      There is also, we now accept, a related wider social value achieved by delineating the proper ambit of the power of the state. The finality involved in the rule against double jeopardy (known by almost everyone, even if not by name) represents an enduring and resounding acknowledgement by the state that it respects the principle of limited government and the liberty of the subject. The rule against double jeopardy is, on this view, a symbol of the rule of law and can have a pervasive educative effect. The rule serves to emphasise commitment to democratic values. As Paul Roberts expressed it:

Double jeopardy protection is very imperfectly expressed in terms of fairness to the accused … It is more illuminating to think of double jeopardy as forming one, significant strand of the limits on a state's moral authority to censure and punish through criminal law. A defendant is not pleading unfair treatment qua criminal accused when invoking the pleas in bar, but rather reminding the state – as the community's representative, the community in whose name the business of criminal justice is done – of the limits of its power. … Defendants asserting double jeopardy protection act almost as private attorneys general, policing the boundaries of legitimacy in criminal law enforcement, keeping state power in check for the benefit of all who value democracy and personal freedom. This is the special value of finality in criminal proceedings, and the principal rationale underpinning double jeopardy protection. The fundamental nature of the values at stake explains why English law's pleas in bar operate as near-absolute barriers to re-prosecution whenever their conditions precedent are satisfied.^[9]

4.18      It is, of course, always the case that the law (and particularly the criminal law) should represent the prevailing values of society, and it is important to recognise that such values can and do change. Even so, double jeopardy serves to maintain confidence in the criminal justice system in a way that is too easily underestimated. The reaction to a particular case can be vocal, powerful and immediate. In a highly charged atmosphere which might understandably arise it may be all too easy to discount the reassurance gained by reflecting, in less emotive circumstances, on long-standing traditional bulwarks of individual liberty.
4.19      We concluded in CP 156 that the rule against double jeopardy was a fundamental principle which protects the individual from oppression by the state and gives the community as a whole a collective sense of security. It reflects one of the principal terms of the bargain between the people and the state, on the basis of which the state has and uses its coercive powers with the consent of the citizen. As a result of the responses to CP 156, and our own reconsideration in the light of those responses, we have come to the conclusion that the principle is even more important than our provisional proposals recognised. In particular, the concept of finality of issues as between the individual and the state is a fundamental process value, which operates on the collective as well as the individual level.

Implications of giving a higher value to finality

4.20      Our conclusion that we significantly undervalued finality has caused us to reconsider the balancing exercise which we undertook. Adopting the language of the analysis in CP 158,^[10] a new evidence exception would have a very high value in terms of accuracy of outcome in those cases to which it applied. If we were to consider only the particular case to which the exception applied, this very high enhancement to the accuracy of outcome might well overcome the process value of finality enjoyed by the individual defendant. We now recognise, however, that we have to consider where lies the balance of advantage between the contending values over the system as a whole. Although the accuracy of outcome value of the exception is very high in those cases in which it is used, if one looks at the system as a whole, the point at which the balance of advantage in favour of reopening acquittals arises, if it ever does, will be different from that identified in CP 156.
4.21      Our judgment is that once greater value is given to finality we can no longer justify an exception as wide in scope as that proposed in CP 156. By "scope", we mean the types of cases which would potentially fall within the exception – in practice, how serious a case has to be, and how probative the new evidence, before the exception can apply to it.
4.22      That does not necessarily mean that no exception can be justified. Any exception must, however, be limited to those types of case where the damage to the credibility of the criminal justice system by an apparently illegitimate acquittal is manifest, and so serious that it overrides the values implicit in the rule against double jeopardy. The boundaries of any such exception must be clear cut and notorious. Thus the question whether there should be an exception at all is inextricably bound up with the scope of any exception. Is it possible to identify a category of cases in respect of which the objective of achieving accurate outcomes clearly outweighs the justifications underlying the rule against double jeopardy?

Defining the category of cases to which an exception might apply

4.23      There are a number of ways in which we might seek to define such a category. At one end of the spectrum, we could do so by venue – by including all indictable offences, or all offences which are triable only on indictment. Although some respondents (including the CPS and ACPO) suggested one or other of these approaches, or variants of them, we consider that it is far too broad a category of case to be justifiable in terms of the principles we have outlined.
4.24      Another possibility is the provisional proposal in CP 156, defining the category by reference to the sentence likely to be imposed on the facts of the individual case. This approach was widely rejected by respondents, including many who supported the broad thrust of the rest of the paper. It was seen as too arbitrary and inexact, and as placing the judge hearing the application in a difficult position. The information available to that judge would be inadequate for a proper, though speculative, sentencing exercise. It would be difficult to assess personal mitigation, and there would be the potential for injustice in relying on the untested prosecution case.
4.25      Further, if the likely sentence were to be the criterion, the great majority of respondents (including some of the prosecutors and police bodies) thought that our proposed minimum of three years' imprisonment was far too low.
4.26      We accept these criticisms. The criterion we proposed would have been too hard to apply, and it set the level of seriousness too low.
4.27      Some respondents suggested more objective measures. The majority favoured determining seriousness by reference to either the type of offence alleged or the maximum sentence available for it. Suggestions included limiting the exception to offences carrying a maximum of 14 years' imprisonment, or to murder alone.
4.28      Some, including the Home Affairs Select Committee, proposed that only offences punishable with life imprisonment should be subject to the exception. In our view this is too blunt an instrument. There is an extensive list of offences which attract the penalty of life imprisonment. It includes some which, although they clearly justify life imprisonment for the most serious instances, also cover comparatively trivial conduct. Robbery, for instance, may amount to theft accompanied by any degree of force, however minor, and may attract any sentence from life imprisonment to a supervision order. In any event, it would be necessary to edit the list. It includes those common law offences for which the penalty is still "at large", and which can therefore technically attract a life sentence. Some of these offences (such as keeping a disorderly house, or blasphemous libel) are clearly not sufficiently serious to justify a life sentence. On the other hand we could not simply exclude all offences in respect of which sentence is at large, because some of them (such as kidnapping) probably could justify a life sentence.

Our approach

4.29      The approach we have decided to adopt, therefore, is to see whether we can identify specific offences, within the larger category of offences potentially attracting a life sentence, which we believe are inherently serious enough to justify the application of a new evidence exception. We have come to the conclusion that, under the present law,^[11] the only such offence is murder.^[12]
4.30      The main reason for this conclusion is the widespread perception, which we share, that murder is not just more serious than other offences but qualitatively different. The effect of this difference is that murder satisfies the test we have proposed^[13] for the scope of any new exception, namely whether a manifestly illegitimate acquittal sufficiently damages the reputation of the criminal justice system so as to justify overriding the rule against double jeopardy.
4.31      As Professor Andrew Ashworth has put it:

The harm caused by homicide is absolutely irremediable, whereas the harm caused by many other crimes is remediable to a degree. Even in crimes of violence which leave some permanent physical disfigurement or psychological effects, the victim retains his or her life and, therefore, the possibility of further pleasures and achievements, whereas death is final. This finality makes it proper to regard death as the most serious harm that can be inflicted on another …^[14]

4.32      This position has been taken in relation to other debates relating to the treatment of murder.^[15] For instance, in its 14th Report, the Criminal Law Revision Committee said:

Should [murder] continue to be regarded, as it has been since the beginnings of our law, as a crime standing out from all others? In our opinion it should. In modern English usage the word "murderer" expresses the revulsion which ordinary people feel for anyone who deliberately kills another human being.^[16]

4.33      In our written evidence to the House of Lords Select Committee on Murder and Life Imprisonment in 1989, we justified maintaining murder as a separate offence, rather than adopting a single offence of homicide:

The Commission sees force in the view that the law should recognise and reinforce the social inhibitions against deliberate killing by placing such killing in a separate category of criminality, and by providing that deliberate killing should only not be murder in specific and specifically defined cases. Thus in law, as in morals, deliberate killing will be marked off as a special category of act, indulgence in which invites condemnation and requires justification of a special kind.^[17]

4.34      On that basis murder, as the most serious form of homicide, is in a unique position and can as a matter of principle be separated off from all other offences. There is, we must accept, a potential problem with using such an analysis to justify treating murder in this way. Murder, as defined in English law, is not confined to deliberate killing. The mental element which must be proved on a murder charge can be satisfied by proof of an intention to cause grievous bodily harm rather than death.^[18] This point is of considerable practical importance. A committee chaired by Lord Lane thought in 1993 that "only a minority of people convicted of the offence have [an intent to kill]".^[19] The result of this, according to the Committee, was that "There is probably no offence in the criminal calendar that varies so widely both in character and in degree of moral guilt as that which falls within the legal definition of murder".^[20] Even if most murderers are not "deliberate killers", however, there is still an important sense in which deliberate killing is the core of the offence of murder. That is sufficient, in our view, to justify concluding that it remains a unique offence, not merely one which is more serious than others.
4.35      Confining the exception to murder would meet our requirement that its scope should be clear cut and notorious. By radically reducing the number of acquitted defendants to whom the exception could ever apply, it would also reduce the number who might be subject to a continuing fear of their acquittals being reopened. It is a striking fact, moreover, that all of the factual or near-factual^[21] concrete examples of cases in which it has been suggested that the new evidence exception could be used have been cases of murder.
4.36      For these reasons we conclude that a new evidence exception is appropriate and desirable in the case of murder.

Manslaughter

4.37      We do not believe that all homicide offences are serious enough to justify bringing them within what is intended to be a narrow exception overriding the principles underlying the rule against double jeopardy. We have concluded that many cases of manslaughter are not so serious as to justify inclusion in such an exception. Yet, we would accept that some forms of manslaughter can involve a comparable degree of moral culpability.
4.38      In our report on involuntary manslaughter,^[22] we pointed out that one of the main drawbacks of the existing offence is its enormous width. The most serious examples border on murder, whereas the least serious are barely more than actionable negligence. We therefore recommended that the single offence of manslaughter be replaced by two new offences, of which the more serious would be called "reckless killing". This offence would be committed by a person who knows that his or her conduct involves a risk of death or serious injury to another, unreasonably takes that risk, and causes death. An example would be the terrorist who places a bomb in a public place, intending to cause panic and chaos without caring whether lives are lost. The offence would be potentially punishable with life imprisonment, and, in our view, would be sufficiently serious to justify applying the new exception to it.
4.39      That recommendation has not yet been implemented, although the Home Office has adopted it in its own consultation paper on the subject.^[23] In view of this we have considered whether to recommend that the substance of the new exception should apply to reckless killing before its creation as a separate offence. This would involve establishing an exception to the rule against double jeopardy where the existing offence of manslaughter is alleged to have been committed recklessly (that is, in such circumstances that it would have amounted to reckless killing had our recommendations for that offence been implemented). We do not believe this would be workable. In practice, prosecutors and courts rarely find it necessary to speculate about whether a defendant's conduct amounted to reckless manslaughter or some lesser form. This would make the criterion very difficult to apply. An alternative approach would be to make the seriousness of the alleged manslaughter a factor for the court to take into account in deciding whether the interests of justice require a retrial. We reject this as being too uncertain and subjective. In cases which are not sufficiently serious to justify allowing the exception to apply, it is in our view better that it should be incapable of applying, rather than leaving it to the court to reject the application as a matter of discretion.
4.40      We therefore conclude that the exception should apply only to murder, until such time as a separate offence of reckless killing is created. If and when that time comes, however, we recommend that the exception should apply to that offence too.

Genocide

4.41      Under the Genocide Act 1969, a person commits the offence of genocide by doing any of a number of specified acts with intent to destroy, in whole or in part, a national, ethnical, racial or religious group, as such. One of the acts specified is the killing of members of the group, and in this case alone the offence is punishable with life imprisonment. In substance this form of the offence is akin to aggravated murder. It would be illogical not to include it in the exception.
4.42      We recommend that the rule against double jeopardy should be subject to an exception in certain cases where new evidence is discovered after an acquittal, but only where the offence of which the defendant was acquitted was murder, genocide consisting in the killing of any person, or (if and when the recommendations in our report on involuntary manslaughter are implemented) reckless killing.

(Recommendation 1)

Retrospective effect

4.43      In CP 156 we invited views on whether any new evidence exception should have retrospective effect, in the sense of applying equally to acquittals which had already taken place before the exception came into force.^[24] On the assumption that (as we now recommend) any new exception would be confined to murder,^[25] the responses were equally divided on this suggestion. Unsurprisingly, those who objected to it included many of those who thought there should not be a new exception at all. Of those who thought there should be a new exception, a clear majority thought it should have retrospective effect.
4.44      We consider that the arguments in favour of giving the exception retrospective effect are powerful. Substantive retrospective criminal legislation renders an act, which was legal when it was performed, subsequently illegal. In the case of the procedural change we propose, the alleged act was already a crime. The new procedure merely makes it possible (or easier) to bring the offender to justice, a desirable outcome whenever it is achieved.
4.45      Further, if the new exception were not retrospective, it could well be a number of years before it could be used. In deciding to recommend a new exception we have taken account of the fact that, in recent years, we have seen considerable advances in forensic science, particularly in DNA analysis. It is the possibility of bringing these new techniques to bear on materials from old cases that is likely to constitute a major source of cases said to fall within the new exception. If there were no retrospective effect, the potential advantage in being able to bring these new techniques to bear on materials from old cases would be lost.
4.46      Furthermore, if the exception were not retrospective, arbitrary distinctions would be drawn between persons who happened to have been acquitted before and after the relevant date. This would open up the prospect of public outrage where new evidence came to light and the exception would otherwise have been available. By recommending that it should be confined to murder, we are limiting the exception to the most serious cases – cases which might be thought particularly to cry out for justice for the deceased and his or her relatives. In such cases, we do not believe that a person against whom there is compelling evidence of guilt should be protected by a mere accident of timing.
4.47      In its report, the Home Affairs Select Committee said:

Retrospection will be a controversial area if legislation is brought forward to amend the double jeopardy law. Without retrospection, the change would take years to have any impact and would leave a sense of frustration about past cases. Time limits would further restrict the benefits of such a change and there is a risk that the strongest cases for a retrial would happen to fall just outside the limits chosen.

We conclude that, if there is a case for relaxing the double jeopardy rule, then it should not be fettered and should apply to past and the future cases without limit.^[26]

4.48      Two members of the Committee expanded on this in the House of Commons debate on the report. Mr Martin Linton MP said:

We think that it should apply retrospectively. As I understand it, the principle is that it would be wrong to change the law so that someone is punished retrospectively for doing something that was not an offence at the time. However, we are talking about people who knew that they were committing crimes, lied in court and got away with it. Such cases are entirely different from those with which the retrospectivity principle intends to deal in law.^[27]

4.49      Mr Paul Stinchcombe MP added:

I am always slow to support retrospective enforcement, but I can conceive of nothing more self-evidently appropriate than where the sole purpose of the legislation in question is to prevent past miscarriages of justice. The double jeopardy rule has been an integral part of our criminal legal system for many centuries, but the time has now come to relax it in order better to protect the integrity of the system and of the citizens of this country.^[28]

4.50      Mr Robert Marshall-Andrews MP, on the other hand, argued that

to make the operation of the statute retrospective would be abhorrent to all criminal concepts, and probably contrary to the Convention.^[29]

4.51      If retrospective effect would infringe the ECHR then it would of course be out of the question; but we do not think it would. Article 7 provides:

No one shall be held guilty of any criminal offence on account of any act or omission which did not constitute a criminal offence under national or international law at the time when it was committed. Nor shall a heavier penalty be imposed than the one that was applicable at the time the criminal offence was committed.

4.52      The first sentence of Article 7 prohibits the creation of retrospective offences by legislation, or through the development of the common law, so as to encompass conduct which would not previously have been regarded as a crime.^[30] It also "embodies, more generally, the principle that only the law can define a crime … and the principle that the criminal law must not be extensively construed to an accused's detriment".^[31] The objective of this guarantee is to ensure that a person should be able to judge, at the time of engaging in particular conduct, whether or not it amounts to a crime.^[32] The article does not, however, prohibit retrospective changes in the rules of criminal procedure so as to remove a bar or obstacle to a prosecution.^[33] The requirements of Article 7 are, in our view, satisfied if the conduct in question constituted a crime at the time when the offence was committed: it is immaterial that the procedural rules in existence at the time of an acquittal or conviction prevented it from being reopened. Article 7 would not prevent the reopening of such an acquittal or conviction under provisions subsequently coming into force.
4.53      One respondent suggested that the change proposed was not merely procedural because it would impose a potential liability to criminal conviction and punishment on people who are presently immune from it. We respectfully disagree. The crucial question, in our view, is whether the effect of the change in the law is to expose the defendant to greater liability than he or she might reasonably have expected at the time of the alleged offence – not some later time when the defendant has been acquitted of it. In our view the clear answer to that question is that it would not. The defendant's exposure is to being convicted of murder, both at the time of the alleged offence and at the time of the retrial.
4.54      The tainted acquittals procedure was not made retrospective. In our view that is not a persuasive precedent as there are differences of principle between that procedure and the one we propose. The tainted acquittals procedure was rightly made prospective only as it involved a new adverse consequence of committing the relevant criminal offence. Had the tainted acquittals procedure been retrospective, it would have been analogous to a retrospective increase in maximum sentence for the administration of justice offence which triggered the application, no less so whether the acquitted defendant, thereby put at further risk, was the person who committed the administration of justice offence or the beneficiary of the commission of that offence.
4.55      On the other hand, we recognised in CP 156 that acquitted defendants will have organised their lives on the justified basis that they would not be troubled by criminal proceedings a second time. This is a serious concern, the more so in the light of our revaluation of the importance of finality in criminal proceedings. Although, in our view, the arguments in favour of retrospective effect are compelling, we accept the force of this particular concern, and we seek in our recommendations to provide some recognition of their force. Where the new evidence is already in the hands of the authorities at the time when the new exception comes into force, we would expect an application for a retrial to be made with all reasonable despatch or not at all. If the prosecution were unduly dilatory in making such an application, under our recommendations the court hearing the application would be required to take account of that delay in deciding whether to order a retrial.^[34] If, moreover, the court hearing the application thought it would be unjust to reopen the acquittal because the defendant had acted in reliance on the assumption that it could not be challenged, that is an argument which a defendant might wish to advance in order to persuade the court to refuse a retrial on the grounds that it would not be in the interests of justice, or thereafter at the retrial to seek to persuade the court to stay the proceedings as an abuse of process. We do not suggest that the date of the acquittal should be wholly disregarded, but only that there should be no absolute bar on retrospective application of the exception.
4.56      We recommend that the new exception should apply equally to acquittals which have already taken place before the exception comes into force.

(Recommendation 2)

What new evidence will trigger the exception?

4.57      In CP 156 we proposed that the prosecution should have to overcome two separate hurdles in relation to the impact of the new evidence on the strength of the prosecution's case.

(1) First it would have to show that the new evidence makes the prosecution's case, as a whole, substantially stronger than it was at the first trial.

(2) Secondly, it would have to show that the likelihood of a conviction at a retrial is of some minimum level. We suggested several possible levels at which the threshold might be set, and invited views on two: namely (a) that it is highly probable that a jury would convict, and (b) that the court is sure that a jury would convict.

A substantially stronger case

4.58      Most respondents supported the first of our proposed hurdles (that the new evidence must make the case substantially stronger than it was at the first trial), but there were notable exceptions. Judge Rodwell QC described the proposal as "a fallacy" because, "as every practitioner knows, no evidence makes a case substantially stronger until that evidence has been tested in court". The Serious Fraud Office thought that the proposed criterion might be difficult to apply in fraud cases. Professor Colin Tapper thought that the application of this criterion would be extremely uncertain and contentious.
4.59      Our proposal was intended to preclude a retrial where the new evidence added only marginally to the cogency of the evidence adduced at the first trial, because in such a case the new evidence might be used as a pretext for what is in effect little more than an appeal against the jury's verdict. The corollary is that this test is more likely to be satisfied where the original case was weak than where it was strong, because where it is weak there is more room for it to be strengthened. We now think that this is a fundamental flaw. The focus of the inquiry should be whether the new evidence is sufficiently strong to justify putting an acquitted defendant at risk for a second time of conviction for the serious offence to which the exception applies, not how strong or weak was the case first time round. We therefore do not recommend the adoption of this criterion.

The likelihood of conviction at a retrial

4.60      The second hurdle we proposed (that the likelihood of the defendant being convicted at a retrial should be assessed as reaching a specified level) attracted a great degree of comment. Nearly all respondents noted the difficulty of arriving at a standard which was appropriately demanding but which did not allow the court hearing the application to usurp the role of the jury at the retrial. The proposed test was also criticised as requiring the court to predict the outcome of a trial which is yet to take place, with the fear that the knowledge that the new evidence has passed such a test would inevitably affect the minds of the second jury. We accept these criticisms. We now believe that the new exception should not require the court to speculate about the likely outcome of a retrial.

Whether the first jury would have convicted

4.61      We have therefore considered possible alternative approaches. One such alternative would be to look backwards to the first trial and ask whether, had the new evidence then been available, the jury would have convicted instead of acquitting. This approach would be comparable to that adopted by the present law in the context of tainted acquittals (where the court must consider how likely it is that, but for the interference or intimidation, the acquitted person would not have been acquitted). We do not think that this approach would be appropriate in the context of new evidence. The tainted acquittal procedure focuses on the legitimacy of the first trial. What happened at the first trial, and what might have happened at the first trial but for the conduct complained of, is of the essence of the exercise. The justification for that procedure is that there has not yet been a proper trial. By contrast, the new evidence exception applies where there has been a proper first trial at which a legitimate verdict was reached. Thus the focus of the question should be whether the effect of the new evidence is such that the first jury's verdict (legitimately reached after a proper trial) cannot in the interests of justice be allowed to stand. What the first jury would, or might, have done if the case presented to it had been different is neither here nor there. Its task is done.

The strength of the new evidence itself

4.62      We therefore think it is necessary to devise a test which does not require the court to speculate either forward or backwards on what a jury would decide in the future, or would have decided in a hypothetical situation in the past, but which focuses instead on the occasion for the application – namely the new evidence itself. In CP 156 we gave examples of the type of evidence which we thought might justify an exception to the double jeopardy rule. The examples we gave were of evidence which was so compelling in itself that, when placed in context, it would have the effect of driving the recipient to the conclusion that the defendant must be guilty. Only in these circumstances do we believe that a retrial should be allowed.
4.63      In CP 156 we considered a test under which the court would judge the independent strength of the new evidence, in isolation from the rest of the evidence.^[35] We rejected this because, save for the possible example of a full and uncontestable confession, any evidence (particularly evidence of a scientific nature) would be meaningless if taken completely out of context. That objection can, however, be met by modifying the proposal. What we now propose is that the strength of the new evidence be considered in the context of the issues that arose at the trial (whether or not matters of dispute between the prosecution and defence), thus enabling its probative force properly to be judged. In this way the strictness of the test is preserved, but it can be applied in a realistic manner. Some examples may helpfully elucidate what we have in mind.
4.64      Where the only issue at trial is identity but after D's acquittal further evidence of D's involvement is discovered (for example, the weapon, bearing D's fingerprints is found in D's house or garden; or DNA evidence showing that D had contact with the victim; or the victim's blood on D's clothing; or CCTV footage showing D at the scene of the crime) it would not be realistic to ask whether such new evidence is in itself compelling evidence of guilt, in isolation from the issues raised at the trial. The probative value of the new evidence will vary according to the matters that are and are not in dispute. For example, it may be the case that in interview or in evidence at the first trial D had denied knowing or having any contact with the victim, or being anywhere near the scene of the crime at the relevant time. Seen in that context, the new evidence may acquire the quality of being compelling. On the other hand D may already have accepted, in interview or in evidence, that he knew the victim, or had been at the scene and come into contact with her and/or was the owner of the weapon (but had lent it to another). In such a case the context within which the new evidence is considered may deprive it of any great significance.
4.65      The proposal does not involve assessing the cogency of the new evidence in the context of the evidence adduced at the trial, except to the extent that it is necessary to examine that evidence in order to identify the issues. Where the issue at the trial was that of identity, for example, the court would have to consider simply whether the new evidence was compelling evidence on that issue – not whether there would be a compelling case in total if the new evidence were added to the old. The point of the exercise is not to consider how strong the original case now is with the enhancement of the new evidence. That would give rise to the risk that cases were reopened merely because there is a bit more to boost what had been a strong case and a surprising verdict. Rather, its point is to enable a case to be reopened when evidence comes to light which is itself so apparently compelling that the court hearing the application is driven to the conclusion that at that stage there is a high probability that the defendant is guilty.
4.66      The test recommended by the Home Affairs Select Committee was whether the new evidence makes the acquittal "unsafe". It thus harks back to the initial verdict. As we have explained, we do not think this is the right approach. In any event, we find the concept of an unsafe acquittal a difficult one. The word "unsafe" is presently used as the test for appeals against conviction. A conviction is unsafe if, upon appeal, there is (or may be at a retrial) reasonable doubt about the defendant's guilt, because, if there were such a doubt, the defendant is entitled to an acquittal. This is the corollary of the rule that the onus is on the prosecution to prove guilt beyond reasonable doubt. Given that rule, however, it is hard to see how an acquittal might properly be described as unsafe.
4.67      Our recommendation is that a successful application to quash the acquittal on the ground of new evidence must satisfy the test that the new evidence is such, when taken in the context of the issues at the trial, that it appears at that stage to be compelling, in that it drives the court to the conclusion that it is highly probable that the defendant is guilty.

Reliability

4.68      One aspect of the cogency of the new evidence is its reliability – the extent to which it is likely to be believed. In the case of oral testimony this may have to be assessed through cross-examination, and in the light of any admissible evidence which may cast doubt on the new evidence or discredit the witness giving it. In the case of scientific evidence it would involve issues such as continuity and the absence of any opportunity to tamper with the evidence. Whilst logically the reliability of the evidence is an aspect of its probative strength, we believe that it would be better to emphasise the importance of this aspect by requiring the court to make a specific finding that the new evidence appears to it to be reliable.
4.69      We recommend that

(1) the new exception should be available only where the court is satisfied that the new evidence

(a) appears to be reliable; and

(b) when viewed in context, appears at that stage to be compelling;

(2) the context in which the court views the new evidence for this purpose should comprise the issues that arose at trial, whether or not a matter of dispute between the prosecution and the defence;

(3) the court should be permitted to have regard to the evidence adduced at trial solely for the purpose of identifying those issues and assessing the impact of the new evidence in the light of them; and

(4) the new evidence should be regarded as compelling if, in the opinion of the court, it makes it highly probable that the defendant is guilty.

(Recommendation 3)

Private prosecutions

4.70      In CP 156 we queried whether a less demanding standard should apply where the previous prosecution was a private one.^[36] Respondents were unanimous that it should not. As one prosecutor noted, a lower standard might result in an increase in private prosecutions. We make no such recommendation.

The interests of justice

4.71      In CP 156 we said we thought it clear that, before quashing an acquittal on the grounds of new evidence, the court should be required to consider whether it is, in all the circumstances, in the interests of justice to do so. For example, it may be clear that for some reason (such as prejudicial publicity, or the lapse of time since the alleged offence) a fair trial is now impossible. By rejecting the application to quash the acquittal in such a case, the court can anticipate the application to stay the proceedings as an abuse of process which would otherwise be made at the retrial. (Granting the application to quash the acquittal, on the other hand, would not preclude an application at the retrial to stay the proceedings on the grounds of abuse of process.) The great majority of respondents were in favour of such a criterion, and we recommend that it be adopted.
4.72      There are certain factors which we believe the court should be expressly directed to consider, in determining where the interests of justice lie. Some of these factors appeared in CP 156 as possible conditions precedent to the quashing of an acquittal. We now consider, however, that they are better treated as factors relevant to the exercise of the court's discretion.

Due diligence in the original investigation

4.73      In CP 156 we proposed that the power to reopen an acquittal on grounds of new evidence should be available only where that evidence could not, with due diligence, have been adduced at the first trial. Although the majority of respondents agreed, we find the arguments of the minority persuasive. There are four principal arguments against a due diligence test.
4.74      The first argument against a due diligence test is that it would not (as we suggested)^[37] work as a disciplinary mechanism at the level of the police, to ensure proper investigation first time. Thus Jeremy Roberts QC asked:

How realistic is it to suppose that the existence of a "due diligence" restriction on the power to receive new evidence would make any significant difference to the efficiency and thoroughness with which the original investigation is carried out (especially in the type of case where it is proposed to give the courts power to reopen an acquittal: by definition these will be cases of serious crime and/or other public importance)? Is it really to be supposed that police officers, customs officers or CPS officials will say: "We won't bother pursuing this line of enquiry now because, if we don't prove our case first time round, we can always have another bite at the cherry"? I would have thought not.

4.75      Others assert the contrary, including some police respondents (such as the RUC). The Jeremy Roberts line is strengthened, vis-à-vis the voices on the other side, by the limiting of the exception to murder, as suggested in his parenthesis. It is also reinforced by Paul Roberts' discussion, which makes the point that there are competing incentives operating on the police, some pushing them towards effective investigation and others pushing them in the other direction, and that, compared to these powerful forces operating in every case, the impact on their behaviour of the potential use of a double jeopardy exception would be small.^[38]
4.76      The second argument against a requirement of due diligence is that it will not work as a disciplinary device at the level of the courts. It is fair to say that this argument was put mostly by those who were against the new evidence exception generally, who thought that there was a danger that an exception would lead to sloppy police work, but did not see our proposal as an adequate safeguard. The Criminal Bar Association, for instance, put it like this:

The "due diligence" test will be easily satisfied … A previously undiscovered witness or unexamined item of clothing or implement may result in significant new evidence. It will not be difficult for prosecutors to assure the court that there was no reason to think X was an eye witness or that an item of clothing belonged to the defendant and hence pass the due diligence test. The acceptance by the courts of routine assurances by the prosecution as to due diligence in custody time limits extension cases (s 22 Prosecution of Offences Act) therefore gives no grounds for thinking this will be a rigorous test.

The force of this argument is that, even if the police would be chastened by the proper application of a requirement of due diligence, it would not be properly applied by the courts.

4.77      A third argument is that it is simply wrong to refuse to reopen a case which should otherwise be reopened, merely because of some extraneous earlier failing of the police. This is always an inherent problem in the use of a procedural mechanism to discipline the police. The sanction is not aimed in any direct sense at the police, but at the public interest in convicting the guilty and at the particular interests in justice of the victim or his or her family. The system imposes a sanction on X to make Y do its job properly. Even if it works, it is hard on X. Thus the House of Commons Home Affairs Select Committee's view that "If a second trial is ruled out because police incompetence had failed in the first, the only winner is the guilty. The victim or the victim's family are left doubly denied justice." The Committee concluded that there should be no due diligence test.
4.78      The fourth argument relates to the process required by our original proposal. In a very interesting discussion of the history of the power of the Court of Appeal to receive new evidence on appeals against conviction, Jeremy Roberts emphasises the long-standing reluctance by the court to use its powers to act on such evidence. One of the effects of this, he says, is to divert the appellant into making criticisms of trial counsel, to get round the objection that the evidence could, and therefore should, have been adduced at trial, whereas the proper focus should be on whether or not the conviction should be upheld. Our original proposal would have a similar effect, diverting attention away from the merits of the case and towards a close examination of the adequacy of the original investigation, even more so than it is presently in appeals against convictions. First, it would be a mandatory requirement, whereas that governing new evidence for the defence is not. Secondly, the competence under examination in defence appeals is mostly that of counsel, or sometimes defence solicitors, which is obviously something that the members of the court are in a good position to assess. In cases of reopening, the focus is more likely to be on the details of the police investigation itself, which may well require expert evidence. It would be unattractive (and we put the argument no higher) if the reopening of an acquittal on a murder charge were to depend on the court's assessment of the competing claims of police experts as to the proper conduct of an investigation which might have taken place some time earlier.^[39]
4.79      These arguments are to a considerable extent compelling (the first three perhaps more so than the fourth), and we now believe that we were wrong to propose in CP 156 that the exercise of due diligence should be a condition precedent for the reopening of an acquittal.
4.80      On the other hand we think it would be wrong to conclude that, far from being a complete bar to such reopening, a want of due diligence should be wholly disregarded. That such a rule would have some impact on the conduct of investigations cannot be overlooked. The fact is that, unfortunately, a combination of resources and human weaknesses already means that some very serious investigations are botched even where there is an absolute rule against double jeopardy. If it is the case, then, that some investigations, even into the most serious crimes, are already conducted lazily or carelessly, it takes little imagination to see that the tendency towards such an approach cannot but be encouraged if, at the back of the mind of the investigator, is the prospect that there may be a second chance and no risk that it will be ruled out for want of diligence, or, perhaps just as importantly, no obligation to have to justify the evidence's non-availability in court. Thus if the investigation is proceeding well but there is a further step which will advance it yet further but which is onerous or expensive, in the absence of any due diligence test at all there may be a temptation, whether conscious or unconscious, not to undertake that further step. In such a case, if, against expectation, the defendant is acquitted, the further step can always be taken as the basis for an application for a retrial. Our view is that these are real concerns against which some safeguards should be erected. They are, however, somewhat remote and to a degree speculative, and do not warrant the erection of a mandatory hurdle.
4.81      Similarly, the argument at the level of the courts is sound, but must not be taken too far. It may well be that a due diligence condition would be less than perfect because some judges may have an unnecessarily unquestioning attitude to bland assurances given by the prosecution. That would not render it wholly useless. The Criminal Bar Association may be right that some judges are too indulgent towards applications to extend custody time limits. We doubt, however, that this failing is so widespread that the protection given to defendants by the custody time limit regime is rendered illusory.
4.82      A discretion to take account of want of due diligence would also have the considerable advantage of broadly mirroring the present state of the law in respect of applications for the admission of new evidence by appellants against conviction in the Court of Appeal. In determining whether to admit such evidence, the court is required to have regard to whether there is a reasonable explanation for the failure to adduce the evidence at the trial^[40] – that is, whether it could, with reasonable diligence, have been obtained for use at the trial.^[41] Additionally, it would arguably be a breach of the principle of equality of arms, and therefore of Article 6 of the ECHR, if the defence could be prevented from adducing new evidence in an appeal against conviction on the ground that the evidence could have been adduced at the trial, but there were no equivalent discretion to prevent the prosecution from adducing new evidence on the same grounds.
4.83      Our judgment, therefore, is that the balance of argument is against a mandatory requirement, but in favour of a statutory direction that the court should take account of any want of due diligence in determining where the interests of justice lie.
4.84      The practical effect of including this as a relevant factor would probably be limited. If the case satisfies our criteria in other respects, it is perhaps unlikely that the court would refuse a retrial solely on the grounds of a want of due diligence in the original investigation. As in the case of fresh evidence relied upon by the defence in an appeal against conviction, the want of due diligence will often be of marginal importance by comparison with the other considerations involved; but in certain borderline cases it may be right that it should tip the balance. The Court of Appeal is well used to applying this factor in the context of defence appeals.

Reasonable despatch since the discovery of the new evidence

4.85      Similarly, we think that any failure by the prosecution to act with reasonable despatch after the new evidence is discovered (or after it would, given due diligence, have been discovered)^[42] is another factor to be taken into account in deciding whether a retrial is in the interests of justice.

Lapse of time since the alleged offence

4.86      In CP 156 we discussed whether the exception for new evidence should be subject to any time limit. We came to no conclusion.
4.87      A narrow majority of respondents were against the idea of a time limit. It was argued that such a limit would arbitrarily exclude cases which were otherwise wholly suitable for retrial. The analogy with war crimes was pointed out. Perhaps the most compelling argument, raised by a number of respondents, was that, where the acquittal took place a long time ago, it may be strongly arguable that a retrial would be an abuse of process. In that case the court could not certify that a retrial would be in the interests of justice.
4.88      It is noteworthy that, of the minority of respondents who were in favour of the idea, some suggested that the time limit should be calculated according to a sliding scale, with the longest time limit (or no limit at all) applying to murder. Obviously the argument for a time limit is weaker if, as we recommend, the exception is in any event to be confined to murder.^[43]
4.89      Our view is that there should be no absolute time limit, but that the length of time which has elapsed since the alleged offence should be another factor to which the court is directed to have regard in determining whether a retrial would be in the interests of justice.
4.90      We recommend that a retrial should be allowed on grounds of new evidence only where the court is satisfied that, in all the circumstances of the case, it is in the interests of justice; and that, in determining whether it is so satisfied, the court should be required to have regard to

(1) whether a fair trial is likely to be possible;

(2) whether it is likely that the new evidence would have been available at the first trial if the investigation had been conducted with due diligence;

(3) whether the prosecution has acted with reasonable despatch since

(a) the new evidence was discovered (or would, with due diligence, have been discovered), or

(b) the new exception came into force, whichever is the later;

and

(4) the time that has elapsed since the alleged offence, together with any other considerations which appear to the court to be relevant.

(Recommendation 4)

The appropriate court

4.91      In CP 156 we proposed that the court charged with determining whether an acquittal should be quashed should in the first instance be the High Court, with a right of appeal to the Criminal Division of the Court of Appeal against a decision to order a retrial.^[44] The majority of respondents agreed with the proposal, but some suggested that the Court of Appeal would be a more appropriate forum to hear the application in the first instance. We find this suggestion persuasive. The Court of Appeal presently considers and decides questions of the introduction of new evidence and whether there should be a retrial on appeals against conviction. Under section 23 of the Criminal Appeal Act 1968 the Court of Appeal has the power to hear any new evidence proposed to be relied on. We anticipate that the statutory provisions governing an application to reopen an acquittal on the grounds of new evidence will give the court hearing the application a similar power. If the court decides to hear the new evidence, it will be desirable for the court which finally decides the matter to be the court which sees and hears the witnesses. Yet further, these decisions may be extremely high profile and controversial, especially if the decision is to reject the evidence or refuse a retrial. In such a case, which depends on judgments on matters of fact, it may be that a decision of a panel of senior judges would be more likely to be, and be perceived as being, correct than that of a single judge.
4.92      If the application for a retrial is first heard by the Court of Appeal, there is no suitable route for an appeal against the granting of the application.^[45] However, if the initial decision is taken at a suitably senior level, a right of appeal is in our view unnecessary. Indeed, if the decision is based on a judgment as to the reliability of the evidence, it might be positively unsatisfactory for an appeal to be heard by a court which had not seen the witnesses. The alternative would be for the witnesses to give their evidence before the Court of Appeal as well as before the first instance court. This would both protract the procedure and subject the witnesses to repetition of the ordeal of giving evidence before even getting to the retrial. The absence of an appeal would also accelerate what might otherwise be a lengthy procedure.
4.93      We recommend that the court empowered to quash an acquittal on grounds of new evidence should be the Criminal Division of the Court of Appeal, and that there should be no right of appeal against that court's decision.

(Recommendation 5)

Evidence which was inadmissible at the first trial

4.94      In CP 156 we proposed that, for the purposes of the new exception, evidence should count as new evidence if, having been inadmissible at the first trial, it becomes admissible through a change in the law. This proposal was comprehensively rejected by respondents, largely through fears that the law might be changed in order to secure a second trial. Even if this seems a little far-fetched, anyone arguing for a change in the law of evidence would be bound to point to examples of cases in which the change would have been effective to secure a conviction; if the argument was successful and the law was changed, the "example" case could be reopened and the effect would be much the same. We consider these objections well founded. We recommend that it should not be possible to apply for a retrial on the basis of evidence which was in the possession of the prosecution at the time of the acquittal but could not be adduced because it was inadmissible, even if it would now be admissible because of a change in the law.

(Recommendation 6)

Successive retrials, and successive applications for retrials

4.95      In CP 156 we proposed that the exception for new evidence should not be available where the acquittal was at a retrial which itself was held by virtue of that exception. A substantial majority of respondents agreed. A further suggestion advanced by some respondents was that there should only be one application for a retrial on this ground, even if the first application is unsuccessful and no retrial has taken place. We accept this suggestion. Where there has already been a proper trial which resulted in an acquittal, it would in our view be oppressive to subject the defendant to repeated attempts at further prosecution. Once one such attempt has been made, the principle of the finality of criminal process should come to the fore.
4.96      Where the acquittal was at a retrial held on some other grounds, on the other hand (for example because the jury at the first trial had to be discharged, or failed to agree), we proposed that this should not be a complete bar to an application for a retrial on grounds of new evidence, but should be only one factor to be taken into account in determining whether another retrial would be in the interests of justice. This proposal provoked a more mixed response, but we think it strikes the correct balance. It follows that this would be another matter to be explicitly considered by the court in deciding whether a retrial would be in the interests of justice.
4.97      We recommend that

(1)where an acquittal is quashed on grounds of new evidence, and the defendant is acquitted at the retrial, no application to quash that later acquittal on grounds of new evidence should be permitted;

(2)where an unsuccessful application is made to quash an acquittal on grounds of new evidence, no further application to quash that acquittal on grounds of new evidence should be permitted;

(3)where a person is acquitted at a retrial held on some other ground,^[46] it should be possible to make one application to quash that acquittal on grounds of new evidence, but the fact that the acquittal occurred at a retrial should be one of the factors to which the court should be required to have regard in determining whether a further retrial would be in the interests of justice.

(Recommendation 7)

Consent to the making of an application

4.98      We have just recommended that the mere making of an unsuccessful application on grounds of new evidence should bar subsequent applications. This makes it all the more important to avoid inappropriate applications being made. It would be unfortunate if, for example, the family of a murder victim were to bring a private prosecution which failed for want of evidence, and then applied for a retrial on the basis of new evidence which was clearly inadequate to satisfy the criteria we propose, with the result that the CPS could not reopen the acquittal even if compelling new evidence did subsequently arise.
4.99      In our view the way to ensure that applications are not made where there is no real prospect of their succeeding is to impose a consent requirement.^[47] HM Customs and Excise argued that applications should only be made with leave of the Attorney-General, and that leave should only be given where the Attorney is satisfied that (i) there is a strong public interest to be served in allowing the application to proceed and (ii) the High Court is reasonably likely to quash the acquittal. We adopt the suggestion, but would substitute the Director of Public Prosecutions (DPP) for the Attorney-General on the grounds that the DPP has primary responsibility for assessing the strength of the evidence against a defendant. In view of the sensitivity of such decisions, however, we suggest that this should be an exception to the usual rule that the powers of the DPP may be exercised by any Crown Prosecutor.^[48] We recommend that it should be necessary to obtain the consent of the Director of Public Prosecutions, in person, before making an application for an acquittal to be quashed on grounds of new evidence.

(Recommendation 8)

Reporting restrictions

4.100      Given that, under our recommendations, a retrial could be ordered only where the new evidence is apparently "compelling" in the sense we have described, there is an obvious danger of prejudice if the jury at the retrial becomes aware of the circumstances in which the retrial was ordered. There is no difficulty merely because the second jury may learn that the trial is a retrial. This often happens at present (for example, where a witness is cross-examined about an inconsistency between the evidence given by the witness at the first trial and that given at the second), and does not appear to present a problem. What the second jury would not learn, unless the defence chose to tell it, is the potentially prejudicial fact that the retrial is being held after an acquittal has been quashed on grounds of new evidence, as distinct from the existing grounds for a retrial.
4.101      Merely not telling the jury that the retrial is being held on grounds of new evidence after a quashed acquittal, however, does not necessarily ensure that the jury will remain unaware of that fact. In a high profile case some of the jurors may know it anyway. The likelihood of their knowing it will be a factor for the court to consider in determining whether a fair trial is likely to be possible, and therefore whether the interests of justice require a retrial. This is not to say that the exception could never be invoked in a high profile case. The risk of prejudice may be effectively countered by a suitable direction to the jury, after consultation with counsel. In the recent case of Montgomery v HM Advocate^[49] the Privy Council expressed the view that seeing and hearing the witnesses may be expected to have a far greater impact on the minds of the jury than such residual recollections as may exist about reports of the case in the media. Further, their impact can be expected to be reinforced by such warnings and directions as the trial judge may think it appropriate to give. The entire system of trial by jury is based upon the assumption that the jury will follow the instructions of the trial judge, and will return a true verdict according to the evidence.
4.102      Even so, we think it desirable that steps should be taken to minimise the risk of the jury becoming aware of the grounds on which the retrial was ordered. There are already a number of provisions prohibiting reporting which might prejudice a subsequent trial. In the case of preparatory hearings, for example, and appeals in relation to them, there is an automatic ban on reporting unless the court orders otherwise.^[50] Clearly a similar provision would be justified in relation to the new exception.
4.103      We recommend that

(1) there should be a prohibition on the reporting of the hearing of an application for a retrial on grounds of new evidence until the application is dismissed or any retrial has finished; but

(2) the Court of Appeal should have power to make an order disapplying or varying that prohibition if

(a) the defendant does not object to the making of such an order, or

(b) having heard representations from the defendant, the court is satisfied that it is in the interests of justice to make it.

(Recommendation 9)

4.104      In addition, the defendant would be free to argue at the retrial that further proceedings should be stayed as an abuse of process, on grounds of prejudicial publicity or any other ground, and whether or not the arguments adduced had already been considered by the Court of Appeal when considering whether a retrial would be in the interests of justice. Ý
Ü   Þ

Note 1    Paras 5.6 – 5.16.    [Back]

Note 2    Para 5.17.    [Back]

Note 3    Paras 5.18 – 5.60.    [Back]

Note 4    “Rethinking Double Jeopardy” [2000] Crim LR 933, 945.    [Back]

Note 5    Professor D W Elliott.    [Back]

Note 6    Paras 4.9 – 4.10.    [Back]

Note 7    Paras 5.14 – 5.15.    [Back]

Note 8    Para 5.14.    [Back]

Note 9    “Acquitted Misconduct Evidence and Double Jeopardy Principles, from Sambasivam toZ” [2000] Crim LR 952, 954.    [Back]

Note 10    See para 7.12 below.    [Back]

Note 11    For another case which may arise in the future, see para 4.38 below.    [Back]

Note 12    In “murder” we include the technically separate offence of genocide where it involves homicide: see para 4.41 below.    [Back]

Note 13    See para 4.22 above.    [Back]

Note 14    Andrew Ashworth, Principles of Criminal Law (3rd ed, 1999) p 263.    [Back]

Note 15    There is even a relevant historical precedent in the form of an Act of Henry VII (1487, 3 Hen VII, c1) which expressly permitted a second trial by “appeal” (akin to a private prosecution) after an acquittal on an indictment for homicide only, despite the double jeopardy rule; see M L Friedland, Double Jeopardy (1969) pp 9–10. The Act was motivated, so it says, by “The King remembering how Murders and slaying of his Subjects daily do increase in this Land … and thereby great Boldness is given to Slayers and Murderers”. Although the provision was principally a means of evading procedural problems caused by the year and a day rule, it was used to try defendants again when they were acquitted against the evidence: eg Young v Slaughterford (1709) 11 Mod 217 and 228, 88 ER 999 and 1007, 18 How St Tr 326. The appeal procedure was abolished in 1819 by 59 Geo 3, c 46.    [Back]

Note 16    Criminal Law Revision Committee, 14th Report “Offences against the Person”, para 15.    [Back]

Note 17    Report of the Select Committee on Murder and Life Imprisonment, vol II – Oral Evidence, Part 1 (HL Paper 78–II) p 119.    [Back]

Note 18    This was not a problem in the context of our 1989 evidence, because we proposed limiting the mens rea of murder to an intention to kill.    [Back]

Note 19    Report of the Committee on the Penalty for Homicide (1993) p 19. The Committee was set up by the Prison Reform Trust.    [Back]

Note 20    Ibid, p 21. Note also Lord Hailsham’s characterisation of mercy killing as “almost venial, if objectively immoral”: Howe [1987] 1 AC 417, 433.    [Back]

Note 21    By “near-factual” examples we mean cases in which a defendant was convicted on the basis of evidence which came to light a considerable time after the offence was committed. If there had been a little more evidence available earlier, it is possible that such a defendant would have been tried and acquitted before the really compelling new evidence came to light.    [Back]

Note 22    Legislating the Criminal Code: Involuntary Manslaughter (1996) Law Com No 237.    [Back]

Note 23    Reforming the Law on Involuntary Manslaughter: the Government’s Proposals, 23 May 2000.    [Back]

Note 24    Para 10.16.    [Back]

Note 25    Two police bodies were prepared to support the idea only in relation to the most serious offences.    [Back]

Note 26    “The Double Jeopardy Rule”, Third Report of the Home Affairs Select Committee (1999–2000) HC 190, paras 54–55.    [Back]

Note 27    Hansard (HC) 26 October 2000, vol 355, col 143WH.    [Back]

Note 28    Ibid, col 151WH.    [Back]

Note 29    Ibid, col 139WH.    [Back]

Note 30    X Ltd and Y v UK (1982) 28 DR 77.    [Back]

Note 31    Kokkinakis v Greece A 260-A (1993) at para 52.    [Back]

Note 32    S W v UK A 335-B (1995); C R v UK A 335-C (1995).    [Back]

Note 33    In X v UK (1976) 3 DR 95, for example, the Commission held that there was no violation of Article 7 where the Court of Appeal upheld the applicant’s conviction by reference to an important precedent in the law of evidence which had been decided by the House of Lords after the conviction.    [Back]

Note 34    See para 4.85 below.     [Back]

Note 35    Para 5.33.    [Back]

Note 36    Para 5.45.    [Back]

Note 37    Para 5.46. See also paras 4.11 and 5.16.    [Back]

Note 38    “Double Jeopardy First Principles and the Criminal Justice Deal: a commentary on the Law Commission’s proposals”, (2001) 64 MLR forthcoming.    [Back]

Note 39    The Court of Appeal has in recent times had some difficulties in determining the proper approach to old cases, where standards have improved since the trial (eg King [2000] Crim LR 835). Similar questions would attend the reopening of an old case. Should the original police investigation be judged by the standards of the time or by today’s standards?     [Back]

Note 40    Criminal Appeal Act 1968, s 23(2)(d).    [Back]

Note 41    Beresford (1971) 56 Cr App R 143.    [Back]

Note 42    Or, where the evidence is already available when the new exception comes into force, after that date.    [Back]

Note 43    And reckless killing: see paras 4.37 – 4.40 above.    [Back]

Note 44    We invited views on whether the prosecution should have a right of appeal against a refusal to order a retrial.    [Back]

Note 45    The success of the application is unlikely to turn on an issue of law suitable for consideration by the House of Lords.    [Back]

Note 46    Including a retrial where the defendant has previously been acquitted but the acquittal has been held to be tainted: see para 5.31 below.    [Back]

Note 47    See generally Consents to Prosecution (1998) Law Com No 255.    [Back]

Note 48    Prosecution of Offences Act 1985, s 1(7).    [Back]

Note 49    The Times, 6 December 2000.    [Back]

Note 50    Criminal Justice Act 1987, s11; Criminal Procedure and Investigations Act 1996, s 37. At para 7.143 below we recommend similar restrictions on the reporting of appeals against acquittals.    [Back]