BAILII [Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]

Scottish Law Commission (Reports)


You are here: BAILII >> Databases >> Scottish Law Commission >> Scottish Law Commission (Reports) >> Double Jeopardy [2009] SLC 218 (Report) (December 2009)
URL: http://www.bailii.org/scot/other/SLC/Report/2009/218.html
Cite as: [2009] SLC 218 (Report)

[New search] [View without highlighting] [Printable RTF version] [Help]


    SCOTTISH LAW COMMISSION

    Report on Double Jeopardy

    Report on a reference under section 3(1)(e) of the Law Commissions Act 1965
    Laid before the Scottish Parliament by the Scottish Ministers
    December 2009

    SCOT LAW COM No 218

    SG/2009/102
    EDINBURGH: The Stationery Office
    £xx.xx
    © Crown copyright 2009
    The text in this document (excluding the Scottish Law Commission logo) may be reproduced free of charge in any format or medium providing it is reproduced accurately and not used in a misleading context.  The material must be acknowledged as Crown copyright and the title of the document specified. Where we have identified any third party copyright material you will need to obtain permission from the copyright holders concerned.
    For any other use of the material in this document please write to the Office of the Queen's Printer for Scotland at Admail ADM4058, Edinburgh EH1 1NG or email: licensing@oqps.gov.uk.
    The Scottish Law Commission was set up by section 2 of the Law Commissions Act 1965[1] for the purpose of promoting the reform of the law of Scotland. The Commissioners[2] are:
    The Honourable Lord Drummond Young, Chairman
    Professor George L Gretton
    Patrick Layden, QC TD
    Professor Hector L MacQueen.
    The Chief Executive of the Commission is Malcolm McMillan. Its offices are at 140 Causewayside, Edinburgh EH9 1PR.
    Tel: 0131 668 2131
    Fax: 0131 662 4900
    Email: info@scotlawcom.gov.uk
    Or via our website at www.scotlawcom.gov.uk – select "Contact Us"
    NOTES
  1. For those wishing further copies of this paper it may be downloaded from our website or purchased from TSO (www.tsoshop.co.uk).
  2. If you have any difficulty in reading this document, please contact us and we will do our best to assist. You may wish to note that an accessible electronic version of this document is available on our website.
  3. SCOTTISH LAW COMMISSION
    Report on a reference under section 3(1)(e) of the Law Commissions Act 1965
    Report on Double Jeopardy
    To: Kenny MacAskill MSP, Cabinet Secretary for Justice
    We have the honour to submit to the Scottish Ministers our Report on Double Jeopardy
    (Signed) JAMES DRUMMOND YOUNG, Chairman
    GEORGE GRETTON
    PATRICK LAYDEN
    HECTOR L MACQUEEN
    Malcolm McMillan, Chief Executive
    2 November 2009
    Contents
      Para
    Part 1 Introduction  
    Our remit 1.1
    Structure of the Report and outline of our recommendations 1.3
    Legislative competence 1.6
       
    Part 2 The rule against double jeopardy  
    Retention of the rule against double jeopardy 2.1
    Statutory restatement 2.4
    The core rule 2.8
    A broader principle? 2.19
    Same facts or same acts? 2.20
    Exceptions to the "same acts" test 2.28
    The impact of Schengen 2.36
    Subsequent death 2.37
    Assault and homicide 2.37
    Unintentionally causing death 2.49
    The problem of overlapping and inconsistent verdicts 2.54
    Verdict or sentence? 2.62
    The treatment of foreign verdicts 2.64
    The incidental proof of foreign offences 2.75
    Nullity of first proceedings 2.76
       
    Part 3 Tainted Acquittals  
    Perjury as a taint 3.9
    The test of a tainted acquittal 3.15
    The purpose of having a test 3.17
    The English example 3.21
    The need for a conviction 3.24
    The nature of the test 3.29
    "Miscarriage of justice" 3.34
    The proposed test 3.36
    Interference with witnesses 3.38
    Interference with jurors or judge 3.44
    To which court should the application be made? 3.49
    Fact finding 3.52
    Power to refuse application in the interests of justice 3.53
    Reporting restrictions 3.55
    Limitations on scope 3.57
    By type of proceedings 3.57
    Retrospectivity 3.59
       
    Part 4 Should there be a "new evidence" exception?  
    Introduction 4.1
    Retrial following admission by acquitted person 4.2
    Retrospectivity 4.8
    A general exception for new evidence? 4.12
    Responses to Discussion Paper 4.14
    Arguments in favour of an exception 4.18
    Consistency 4.20
    Balance 4.23
    Scientific and technical advance 4.25
    Public perception 4.33
    Moral principle 4.35
    Reasons for the rule against double jeopardy 4.37
    Conclusion 4.43
       
    Part 5 The formulation of a "new evidence" exception  
    Introduction 5.1
    To what offences might a new evidence exception apply? 5.2
    The test of "newness" 5.7
    New evidence in England and Wales 5.8
    The need for reasonable diligence 5.14
    The effect of changing rules of admissibility 5.16
    The significance of the new evidence 5.21
    The overall strength of the case 5.22
    Substantially improving the changes of conviction 5.24
    "Compelling" evidence 5.29
    "Compelling" evidence in practice 5.32
    "Miscarriage of justice" 5.45
    Conclusions 5.56
    A proposed test 5.60
    The interests of justice 5.63
    Reporting restrictions 5.66
    Retrospectivity 5.71
       
    Part 6 Summary of recommendations
    Part 6
       
    Appendix A:

    Double Jeopardy (Scotland) Bill
    A
       
    Appendix B:

    List of Respondents
    B

    Note 1   Amended by the Scotland Act 1998 (Consequential Modifications) (No 2) Order 1999 (SI 1999/1820).    [Back]

    Note 2   As at 2 November 2009.    [Back]

    Part 1 Introduction
    Our remit

    1.1              On 20 November 2007 we received a reference from the Cabinet Secretary for Justice, Mr Kenny MacAskill MSP, asking us:

    "To consider the law relating to:
    and to make any appropriate recommendations for reform."
    This Report relates to the second part of this reference, namely the principle of double jeopardy and whether there should be exceptions to it. We consider the scope and implications of the rule in some detail in this Report. Broadly, however, it prevents someone from being tried twice for the same offence. The first part of the reference was addressed in our report on Crown Appeals, which was published in July 2008,[1] and we aim to report on the remaining aspects of the reference in 2010 or 2011.

    1.2         We published Discussion Paper No 141 on Double Jeopardy in January 2009 and there followed a three month consultation period during which we received a number of informative and helpful responses.[2] We have had the further benefit of meeting representatives of the Crown Office and Procurator Fiscal Service ("COPFS") to discuss in detail their response to the Discussion Paper; we are grateful for their input.

    Structure of the Report and outline of our recommendations

    1.3         In Part 2 we consider the rule against double jeopardy. We conclude that the present common law is unclear, and recommend that the rule against double jeopardy should be clearly stated in statute. We recommend that there should be a "core" rule, prohibiting a second trial for the same offence as that of which a person has already been convicted, for an alternative offence of which the accused might have been convicted on the original indictment or complaint, or for an aggravated form of the original charge. In addition, we recommend that there should be a broader principle, enforceable by the courts, against multiple trials in relation to the same acts. We recommend that it should continue to be possible to prosecute someone for murder or culpable homicide where he has previously been convicted, prior to the death of the victim, of assault; but that it should no longer be possible to do so where the outcome of the first trial was acquittal. We further recommend that the same rule should apply to statutory offences of causing death (such as causing death by dangerous driving). We also recommend that an acquittal or conviction in a foreign jurisdiction should in principle bar proceedings in Scotland in respect of the same acts, but that the Scottish court should be able to disregard such foreign proceedings where it concludes that it would be in the interests of justice to do so.

    1.4         In Part 3 we consider whether it should be possible to retry an acquitted person where it can be established that the trial was tainted by an offence against the administration of justice. We conclude that the rule against double jeopardy should only apply where there has been a proper trial, and that it should be possible to retry an acquitted person where the first trial was shown to be tainted. We suggest that the question of whether the trial was tainted is independent from that of whether the acquitted person was involved in the tainting offence, and that it should not be necessary, in order to justify a retrial, to prove the acquitted person's involvement in that offence. We recommend that the High Court should have the power, on the application of the Lord Advocate, to set aside an acquittal and grant authority for a retrial where it is satisfied that the original trial has been tainted, and make detailed recommendations regarding what should constitute a taint and how the existence of a taint should be established.

    1.5         Part 4 considers whether there should be an exception to the rule against double jeopardy where, following an acquittal, the acquitted person credibly admits or confesses to having committed the offence of which he was acquitted. It goes on to consider whether there should also be a broader exception, allowing a retrial where new evidence of guilt emerges. We note that while there are a number of powerful arguments in favour of allowing such an exception, there are also powerful arguments against. We are unable to reach a firm conclusion as to whether or not such an exception should be allowed, and make no recommendation. We consider, however, that it would be helpful to make some recommendations regarding the formulation of any new evidence exception. These recommendations are contained in Part 5. Part 6 contains a summary of our recommendations. Appendix A contains our draft Bill, with notes on sections, and Appendix B a list of those who responded in writing to our Discussion Paper.

    Legislative competence

    1.6         The recommendations set out in this Report relate to criminal law. With a few exceptions, which do not concern any of the matters in this Report, this area of law is not reserved to the United Kingdom Parliament and so falls generally within the legislative competence of the Scottish Parliament.[3]

    1.7         The legislative competence of the Parliament is also subject to compliance with the Convention rights and with Community law.[4] Community law, in the form of Article 54 of the Schengen Convention,[5] prohibits a Member State from prosecuting a person in respect of acts which have previously been the subject of prosecution in another Member State.[6] Any Scottish legislation which purported to apply exceptions to the rule set out in that Article would, to that extent, be outside legislative competence. Provided, however, that any legislation implementing our recommendations is framed so as not to apply in cases governed by Article 54, we consider that our proposals are not contrary to Community law.

    1.8         Nor do our recommendations conflict with Convention rights. The possibility of reopening a criminal case after a verdict of acquittal is prima facie compatible with Article 6 of the European Convention on Human Rights.[7] The right not to be tried again for an offence of which one has already been punished is contained in Article 4 of the 7th Protocol to the Convention. This Protocol has not been ratified by the United Kingdom, although the UK government has indicated on a number of occasions its intention to do so. Accordingly, Article 4 of Protocol 7 does not at present constitute a "Convention right" for the purposes of the Scotland Act and therefore does not place any limit on the legislative competence of the Parliament. Nor do we consider that our proposals would conflict with that Article when and if it becomes such a "Convention right".[8]

    1.9         We note the views of the Law Commission of England and Wales in relation to the retrospective application of a tainted acquittal procedure, which they suggested might raise issues of compatibility with Article 7(1) of the Convention as representing the imposition of a more severe penalty for the offence against the administration of justice than applied at the time when the offence was committed. We are of the view that this concern is without foundation, and that the retrospective application of a tainted acquittal procedure would be compatible with the Convention.[9]

    1.10         For reasons unrelated to Convention compliance, we recommend that any new evidence exception should not be made retrospective. We identify a possible issue regarding the compatibility of a retrospective exception on the basis of new evidence with the right to a private and family life in terms of Article 8(1) of the Convention. However, since we make no recommendation in relation to the introduction of a new evidence exception, and recommend against making any such exception as might be introduced retrospective, we do not need to reach a concluded view upon whether there would be merit in an Article 8 challenge to such retrospective application.

    1.11         We note that it seems probable that the Treaty of Lisbon, and therefore the Charter of Fundamental Rights, will come into force in the near future. As we observed in Part 4 of the Discussion Paper, this will bring into force a different range of Community obligations in relation to Double Jeopardy. No doubt the Scottish Government and Parliament will themselves consider the draft Bill in the light of the coming into force of the Treaty, and reach their own conclusions as to the effect of the Treaty on the legislative competence of the Parliament.

    Part 2 The rule against double jeopardy
  4. Retention of the rule against double jeopardy

    2.1         The most basic question to be considered in any review of the law of double jeopardy is whether the rule continues to serve a useful purpose. Perhaps the best reform would be to abolish the rule against double jeopardy in its entirety?

    2.2         In the Discussion Paper we identified three main features of the rule against double jeopardy which, in our opinion, make it indispensable.  First, the rule is a fundamental recognition of the finality of criminal proceedings.  Finality of criminal verdicts provides at least two major benefits to society.  The individuals involved in the trial can continue with their lives confident in the knowledge that the matter has been resolved.  In addition, there is the more general benefit that public confidence in the efficacy of the court system is bolstered.[10]   Second, the rule has an important function in expressing the limits of the power of the state vis-à-vis the citizen.[11]   Third, the rule against double jeopardy affords protection from the anxiety and humiliation that repeated trials would undoubtedly cause accused persons.  Protection from such stress is what Hume described as the "obvious and humane consideration" which is in his view the primary justification for the rule against double jeopardy. These considerations are as valid in modern times as they have ever been. We have therefore concluded that the rule against double jeopardy remains essential to the rule of law.  The overwhelming majority of those who responded to the Discussion Paper agreed with our assessment of the continuing importance of the rule against double jeopardy and with our proposal that it be retained.

    2.3         We recommend:

    1. There should continue to be a general rule against double jeopardy.
    Statutory restatement

    2.4         In Part 3 of the Discussion Paper we summarised the present Scots law of double jeopardy. We noted that some recognition of the finality of criminal verdicts could be found in the earliest written materials of Scots law and that a rule against multiple trials was recognised by the 13th or 14th Century and perhaps significantly earlier. By the close of the 18th century, the rule was sufficiently established to be stated by Hume as a maxim of our law:

    "The prime benefit of a sentence of absolvitor is, that the pannel can never again be challenged or called in question, or made to thole an assize (as our phrase for it is) on the matter or charge that has been tried. The ground of which maxim lies in this obvious and humane consideration, that a person is substantially punished, in being twice reduced to so anxious and humiliating a condition, and standing twice in jeopardy of his life, fame or person."[12]

    2.5         Although it has long been clear that no-one could be tried twice for the same "matter or charge that has been tried", there have been relatively few reported cases in which issues of double jeopardy have arisen, and it is not possible to say with any certainty exactly where the boundaries of the rule lie. What is "the same matter or charge"? The decided cases are not easily reconciled, and provide only the most general indication.[13] Must there have been a sentence imposed before a prior conviction will afford double jeopardy protection to the accused who is again charged with a crime arising out of the same incident? It appears that different rules may apply in solemn and summary proceedings, but again the present law is unclear.[14]

    2.6         In view of the uncertainties in the existing law, we suggested that it should be simplified and restated in codified form. We acknowledged that those very uncertainties meant that any attempt to codify the rule against double jeopardy in statute would almost certainly involve some alteration to the existing law. Respondents to the Discussion Paper agreed that the present law was unsatisfactory and would benefit from restatement and reform.

    2.7         We recommend:

    2. The general rule against double jeopardy should be reformed and restated in statute.
    The core rule

    2.8         Part 6 of the Discussion Paper contained a detailed discussion of a possible basis upon which the rule against double jeopardy might be restated. The discussion reflected the dual purposes served by the rule, namely the promotion of finality in criminal litigation and the avoidance of unnecessary distress to the accused through repetition of the criminal process.

    2.9         We noted that respecting the finality of verdicts will not always be sufficient to protect the accused from the anxiety of further proceedings, since a single set of facts may disclose a number of distinct offences. So, for instance, a person who, while drunk, steals a car and crashes that car into a bus queue may be charged with drunk driving, theft, and dangerous driving. In terms of the finality of verdicts, there would be nothing incompatible in finding the accused not guilty of driving while drunk, but guilty of theft; or not guilty of theft but guilty of drunk driving; or not guilty of either theft or drunk driving, but guilty of dangerous driving, and so on. It would, however, be objectionable without good reason to separate the charges so that each offence was tried in separate proceedings: the Crown should not be permitted, for instance, to hold the theft charge as a backup, to be charged only if it is unsuccessful in securing a conviction on the dangerous driving charge.

    2.10         While the aims of finality and protection of the accused overlap, the finality of the verdict in the first trial can be respected without affording the accused adequate protection against the distressing prospect of further proceedings arising from the same incident. While the finality of the original verdict may be respected by referring to the question of what was litigated and what was decided,[15] adequate protection of the accused from the threat of subsequent prosecution in respect of the same incident requires us to consider not merely what was litigated in the first proceedings, but also what should have been litigated, the general rule being that all charges arising out of the same incident should be brought together wherever possible.

    2.11         "Wherever possible" is an important qualification. It is never appropriate to try someone twice for the very same offence[16] but there are a number of circumstances in which it might be appropriate to prosecute in separate trials different charges arising from the same incident or acts of the accused. The most obvious and uncontroversial case is where it would not be fair to bring all of the charges in the same proceedings. Returning to our example of the drunken driver and the bus queue, it would not have been appropriate to bring a charge of driving while disqualified in the same proceedings as the charge of dangerous driving, since the prosecution of the former charge would have required the disclosure of the past driving convictions which led to the accused's disqualification. More generally, there should be no bar on separately prosecuting different charges arising from the same incident where this manner of proceeding has been agreed in advance between the defence and the prosecution, or where such a separation of charges is necessary, in the interests of justice, to avoid unfairness at one or other of the trials.

    2.12         Another rather more difficult case in which it might be appropriate to prosecute a second time in relation to the same acts of the accused would be where the original charge was later discovered to have related only to one aspect of a larger and more significant crime.

    2.13         After examining the various tests applied in common law jurisdictions and under the Schengen convention, we suggested that it would be appropriate to pursue a dual approach to double jeopardy protection, in which a reasonably narrow core rule against double jeopardy was supplemented by a broader principle against unreasonably splitting cases.[17]

    2.14         We proposed, as the core rule, that:

    A second prosecution should be prohibited where a person has previously been convicted or acquitted of an offence, and:
    (a) the second indictment or complaint charges an offence of which it would have been competent to convict the accused on the first indictment or complaint (so, for instance, an earlier verdict on a trial for murder will bar subsequent prosecution for attempted murder, culpable homicide, assault etc.); or
    (b) the second indictment or complaint charges an offence which is an aggravated form of the offence charged on the first occasion (so, for instance, a previous conviction or acquittal of assault will bar a subsequent prosecution for assault to severe injury).[18]

    2.15         Those who responded to our Discussion Paper were broadly content with our proposed formulation of the rule. One respondent expressed concern that the reformulated rule should not be taken as suggesting that murder is merely an aggravated form of assault; we return to this question below. Another suggested that the core rule should be based not upon a test of the same offence but rather upon prosecution for the same wrong. Whatever the conceptual merits of an analysis based upon the same "wrong", we consider that the limits of the core rule against double jeopardy are better defined in more concrete and familiar terms, referring to the same offence and to offences of which the accused might have been convicted on the original complaint or indictment. Such a test would be relatively straightforward to apply, whereas a test based upon identifying the wrong constituted by each offence would lead to some uncertainty. For example, it is not clear that a rule which prohibited successive prosecutions of the same wrong would generally prevent the successive trials of an offence and its attempt, or of an offence and a lesser included offence. Is assault the same wrong as attempted murder? Reset the same wrong as theft? Sexual assault the same wrong as rape? We think that there is, at the very least, room for doubt; and this is sufficient for us to prefer the more familiar analysis which we have suggested to a novel analysis based upon identifying wrongs.

    2.16         Another perspective was provided by the Faculty of Advocates, who commented:

    "We can envisage a situation where fresh evidence comes to light after conviction, which, had the evidence been available at the trial, might have resulted in conviction for a lesser crime. For example evidence of intervening medical negligence. In such circumstances (after the conviction in relation to the original trial was quashed), a second prosecution for a lesser crime, of which it would have been competent to convict the accused on the first indictment, should be permitted.
    Accordingly, where the original trial resulted in a conviction, we are of the view that a second prosecution should only be prohibited in circumstances outlined in subsection (b) [of the proposed test]: where the second indictment or complaint charges an offence which is an aggravated form of the offence charged on the first occasion." (emphasis added)

    2.17         This comment highlights an important point. Our proposed core rule against double jeopardy is intended to apply only where there is a subsisting verdict; that is, where the first proceedings resulted in a verdict which has not been quashed. It is not intended to prejudice the existing provisions of the Criminal Procedure (Scotland) Act 1995 whereby the High Court, in disposing of an appeal against conviction, may quash the original conviction and grant authority to bring a new prosecution.[19] Once this is made explicit, the Faculty's objection to limb (a) of our proposed test falls away.

    2.18         We recommend:

    3. A second prosecution should be prohibited where a person has previously been convicted or acquitted of an offence and:
    (a) the second indictment or complaint charges an offence of which it would have been competent to convict the accused on the first indictment or complaint (so, for instance, an earlier verdict on a trial for murder will bar subsequent prosecution for attempted murder, culpable homicide, assault etc.); or
    (b) the second indictment or complaint charges an offence which is an aggravated form of the offence charged on the first occasion (so, for instance, a previous conviction or acquittal of assault will bar a subsequent prosecution for assault to severe injury).
    (Draft Bill, section 1(1))
    A broader principle?

    2.19         As the basis for applying a broader principle, we identified two possible types of test. The first test would ask whether the second prosecution concerned the same facts which formed the basis of the first prosecution; the second, whether the second prosecution related to the same acts of the accused.

    Same facts or same acts?

    2.20         We suggested that a "same facts" test had a number of problems. A test which applied only where the facts were precisely the same as those which gave rise to the first charge would offer little protection against inappropriate repetition of proceedings. In the example which we gave in the Discussion Paper,[20] a man who was tried and acquitted of raping his sister could arguably be tried again, under such a "same facts" test, for incest, on the basis that each crime requires proof of a fact that the other does not. Even a test that referred to substantially the same facts could give rise to some surprising outcomes, since there is no principled way of deciding, in the abstract, which facts are relevant. Returning again to the example of the man and his sister, it may be that on such a test the question of whether the second charge arose out of "substantially the same facts" as the first would depend not upon what the accused had done, but rather upon the way that the first indictment was framed and the manner in which the evidence emerged at the first trial. If the first trial did not raise the issue of the blood relationship between the accused and the complainer, then it might well be open to a court to conclude that the "facts" were substantially different.[21]

    2.21         There are at least two Scottish cases which suggest that a "same facts" test would be inappropriate. In Glen v Colquhoun[22] the accused, having once been tried for having been on a stretch of river with intent illegally to take salmon and having in their possession a net and boat with such intent, were again charged with inter alia fishing for salmon with a net having a mesh contrary to relevant byelaws. Lord Ardmillan held that the previous proceedings did not bar the charge relating to the use of the net with the unlawful mesh. Did the charge of fishing with an unlawful mesh arise out of the same facts as the earlier prosecution, namely being on the river with a net and boat having the intention unlawfully to take salmon? We see no clear answer to this question. In Galloway v Somerville[23] the Circuit Court of Justiciary held that charges of poaching and of possessing game without a licence depended "upon entirely different facts", even where the two charges arose from a single occasion upon which the accused was found in possession of a hare. If such an approach were taken to the identification of the "facts" of a case, there would be a real risk that a "same facts" test would sometimes allow a second prosecution in relation to the same conduct of the accused, and in relation to charges which might competently and practicably have been brought together.

    2.22         We suggested that a more attractive approach would be a test that focused not upon whether the second prosecution arose from the same facts as the first, but rather upon whether it concerned the same acts of the accused. We noted that this was the approach which the UK was bound to adopt to those who had already been subjected to criminal prosecution elsewhere in the EU, by virtue of the test developed by the European Court of Justice in applying Article 54 of the Schengen Convention.[24] According to that Court, in assessing what constitute the "same acts", "the relevant criterion . . . is identity of the material acts, understood as the existence of a set of facts which are inextricably linked together, irrespective of the legal classification given to them or the legal interest protected."[25] Such an approach would give substantially more protection against successive prosecution than a narrower "same facts" approach. It would bar the prosecution for incest in our hypothetical example of the brother tried for raping his sister, since regardless of whether the legal wrong is rape or incest, the charges would relate to a single act of the accused (that is, having sexual intercourse with the complainer). It would bar a second prosecution on the facts of Glen v Colquhoun, since the second charge arose out of the same act of fishing as the first (the characteristics of the net itself being merely another respect in which the same act might constitute an offence). It would most likely also bar a second prosecution in a case such as Galloway v Somerville, since while poaching and possessing game without a licence may be seen as separate acts, and are certainly separate offences, it could readily be argued that the alleged act of poaching and the possession of game without a licence "constitute a set of facts which are inextricably linked together in time, in space and by their subject-matter."[26]

    2.23         We observed that it appeared to be the present practice of the Crown to bring all charges relating to a particular incident in the same proceedings wherever possible, and that a rule which generally prevented subsequent prosecutions arising from the same acts of the accused would effectively place this long-standing practice of the Crown upon a statutory footing.

    2.24         Respondents generally supported the introduction of a broader rule, and those who did so generally agreed with our proposal that it should be based upon a "same acts" rather than a "same facts" test. One respondent opposed the introduction of a broader rule, in the interests of clarity. We recognise that any such rule will inevitably be less clear in its application than our proposed core rule as the application of the rule will depend upon the facts of each case. Adopting a broader rule inevitably brings some loss of clarity. We consider, however, that this loss of clarity is a price worth paying in order to place the protection of the citizen against unjustified multiple prosecution upon a sound legal footing.

    2.25         COPFS agreed that the "same acts" test was preferable to a "same facts" test, but suggested that the law already provided the accused sufficient protection in the form of a plea of oppression, as explained by the High Court in HMA v Stuurman.[27] There may be circumstances in which a plea of oppression would be relevant, but the court in Stuurman - a case concerned not with multiple proceedings but with potentially prejudicial publicity - characterised the test of oppression as concerning whether it would be possible for the accused to have a fair trial. In the context of double jeopardy protection, the question is not whether the second trial would, in its own terms, be a fair one, but rather whether it would be fair to have the second trial at all. COPFS acknowledged that our proposal reflects their existing practice and made no objection to this practice being placed upon a statutory footing.

    2.26         Two respondents suggested that it would be desirable to deal with the broader rule against multiple proceedings in relation to the same acts as part of a more general discretion of the court to prevent abuse of process. We see some merit in this, but consider that to recommend the introduction or recognition of a more general abuse of process jurisdiction in the High Court would be beyond our present terms of reference.

    2.27         We recommend:

    4. Beyond recognising a rule against trying a person again for an offence of which he has already been convicted or acquitted, Scots law should recognise a broader principle that a person should not be tried again in relation to the same acts which gave rise to that prosecution.
    (Draft Bill, section 2)
    Exceptions to the "same acts" test

    2.28         Having concluded that our law should recognise a broader principle that a person should not be tried more than once in relation to the same acts, the next question is whether this principle should be expressed as a rule with strictly defined exceptions (for example, where the accused had himself sought a separation of trials[28]), or as a broader judicial discretion. If the latter, should there be a presumption that the second prosecution will be barred unless the Crown can satisfy the judge that there are special circumstances justifying the separation of trials?

    2.29         There was widespread support among respondents for leaving the question of which subsequent charges were barred as a matter for the discretion of the trial judge, having regard to the facts and circumstances of the individual case, with a presumption that subsequent charges arising from the same acts would be barred unless the Crown could satisfy the judge that there were special circumstances justifying the second prosecution.[29] However, one respondent, PW Ferguson QC, sounded a note of caution:

    "[v]iewing the matter as a discretionary power of the trial judge to bar a second prosecution suggests that a trial judge could legitimately regard the second trial as being based on the same acts but nonetheless, in his discretion, repel the plea because, for example, it was in the public interest to try the accused on the second libel. Moreover, if it were a discretionary power the Appeal Court's jurisdiction in any appeal would have logically to be limited to applying the test for discretionary decisions set out in Wordie Property Co Ltd v Secretary of State for Scotland.[30]"

    2.30         COPFS opposed the imposition of an onus upon the Crown to justify the prosecution by reference to special circumstances. They again referred to the availability of a plea of oppression and suggested that it would in any case be difficult to identify exactly what should count as special circumstances justifying a further prosecution.

    2.31         There are two separate questions here. The first is whether the second charge arises out of the same acts as the first. It should be for the accused to raise the issue of the previous prosecution as a plea in bar of trial; and if the accused does not take the point in advance of the trial, and persuade the court that the instant prosecution does indeed relate to the same acts as an earlier trial, the issue of justification does not arise. Once it has been established to the court's satisfaction that the instant prosecution relates to the same acts as an earlier trial, it should be for the Crown to show that there are special circumstances justifying the prosecution.

    2.32         What are "special circumstances"? Mr Ferguson's principal criticism of the proposed discretion, quoted above, was that it would allow the trial judge to conclude that the instant prosecution concerned the same acts as an earlier trial, but that it should nevertheless be allowed to proceed. We do not regard this as a defect in our proposal. There are circumstances in which it would be appropriate for the judge to conclude that a second prosecution concerned the same acts as a first, but that it should nevertheless be allowed to continue. The most obvious and uncontroversial example is the case in which the accused had sought or consented to a separation of charges, for example where one of the charges arising from the acts which formed the subject of the first proceedings would have involved the disclosure of the accused's past convictions, prejudicing the fairness of his trial on the remaining charges. A commonplace example would be a case in which one of the possible charges arising from an incident was of driving while disqualified, the proof of which charge would require proof or admission of the accused's prior driving convictions which led to the disqualification.

    2.33         We have considered whether this uncontroversial exception to a general rule against multiple trials arising from the same acts should be the only exception. Perhaps, rather than referring somewhat vaguely to "special circumstances", the rule should simply be that there may be no second trial of the same acts, except where the accused has consented to a separation of charges or where, despite an absence of explicit consent, to have tried the present charges at the first trial would have been unfair to the accused? While we think that this situation is uncontroversial, and common, enough to warrant its own specific provision in the legislation establishing the broader rule against double jeopardy, we think that there is also room for a broader discretion to allow further prosecution in relation to the same acts in special circumstances.

    2.34         The need for this broader discretion arises from the potential breadth of the "same acts" test and the need to avoid any doubt about the continuing competency of including charges in an indictment or complaint for evidential purposes. Sometimes, in order to prove a particular offence, it is necessary to prove conduct which might itself have been the subject of a criminal charge. Although the conduct is narrated in the indictment (and so is, technically, charged)[31] the prosecutor has no intention of seeking a conviction and includes the other offence only in order to provide a basis for leading evidence relevant to the principal charge. Whether or not such charges, included for evidential purposes, will be regarded as relating to the same acts as the principal charge will depend upon the facts of each particular case, and the degree to which the charges are separated in time, in space and by their subject-matter. Often such charges will not relate to the same acts, but even where they do, the prosecution should not be barred where the court is satisfied that the prosecution of the principal charge is not contrary to the interests of justice.

    2.35         The conduct which justifies a first prosecution for a relatively minor offence may later need to be proved in establishing a more significant offence, and this should not be prevented by an over-rigid rule against successive trials in relation to the same acts. It should be borne in mind that the justification for the broader principle against multiple proceedings arising from the same acts is based on a concern to avoid what would, in substance, be abuses of process - the improper splitting of charges so as to afford the prosecution multiple chances of convicting the accused in respect of a single set of acts. Where there is no such impropriety, and where the public interest otherwise favours prosecution, the court should have discretion to allow further proceedings notwithstanding the fact that they concern the same acts as formed the subject matter of an earlier prosecution.[32]

    The impact of Schengen

    2.36         We should note at this point that the discretion to allow subsequent proceedings in respect of the same acts cannot apply in relation to those who were first proceeded against in another EU state, since Article 54 of the Schengen Convention makes no provision for any such exceptions. This is not surprising. As we have noted, the most common situation in which an exception will be appropriate is where the decision has been taken, with the active consent or at least the acquiescence of the accused, that it would be appropriate in the interests of affording the accused a fair trial for the charges against him to be split and certain of those charges to be made the subject of separate proceedings. Such a situation is most unlikely to arise in the Schengen context, in which more than one state's prosecuting authorities are involved. Any legislation dealing with double jeopardy should nevertheless make it clear that it does not apply to cases governed by the Schengen Convention.

    Subsequent death
    Assault and homicide

    2.37         A rule against multiple proceedings in relation to the same acts has obvious relevance to the question of whether a person tried for an offence of assault may be tried again for murder or culpable homicide if the victim of the assault subsequently dies.

    2.38         We discussed the present law at paragraphs 6.49 to 6.58 of the Discussion Paper. As one would expect, such cases arise infrequently. The case law, such as it is, suggests that a person who has been tried for assault may be tried again for murder if the victim dies after the first trial, regardless of whether the first trial resulted in conviction or acquittal.[33] The reasoning behind this rule was explained by Lord Ardmillan in the 19th century case of James Stewart in a passage which was quoted with approval by the High Court in Tees v HMA:[34]

    ". . . there never can be the crime of murder until the party assaulted dies; the crime has no existence in fact or law till the death of the party assaulted. Therefore it cannot be said that one is tried for the same crime when he is tried for assault during the life, and tried for murder after the death, of the injured party. That new element of the injured person's death is not merely a supervening aggravation; but it creates a new crime."

    2.39         The same result would arise from the application of our proposed core rule against double jeopardy.[35] The accused could not have been convicted of murder at the original trial for assault, since the victim had not then died. As murder is not merely an aggravated form of assault, but a distinct crime, prosecution for murder would not be barred under the second leg of our core rule, which prevents subsequent prosecution for an aggravated version of the crime originally tried.

    2.40         We suggested in the Discussion Paper that, although such a prosecution for murder would be in respect of a different crime, it would nevertheless involve the accused being tried again for precisely the same act.[36] Some respondents to the Discussion Paper questioned the accuracy of this statement. Professor Paul Roberts described our analysis as "wrong, or at least metaphysically inadequate," maintaining that:

    "The issue is not only about the results of D's act, but about what D's act was . . . After V dies the nature of D's act changes (or, what is for present purposes the same thing, what we can legitimately say about D's act changes). By the time of the second trial, all the latter expressions encapsulating the fact that D killed V and (let us stipulate) is consequently guilty of V's murder now become true statements of fact about D's act or conduct, not merely about the results of an act which remains static and complete the moment that D's body has finished moving in time and space."

    2.41         We do not think that it is wrong (or even metaphysically inadequate) to restrict the consideration of the accused's acts for this purpose to his physical actions ("Did A hit V on the head with a shovel?") rather than retroactively characterising the same physical action as a different "act" in light of consequences which become apparent only after the first trial. To do otherwise would be to treat the answer to the question "Did A murder V (by hitting him on the head with a shovel)?" as independent of the answer to the question "Did A hit V on the head with a shovel?", whereas it is obvious that if the answer to the second question is that he did not, the answer to the first question must also be no.

    2.42         All of our respondents agreed that it should continue to be possible to prosecute for murder or culpable homicide where the victim dies after the accused has been convicted of the assault which is alleged to have led to the death. Opinion was divided, however, as to whether it should continue to be possible to prosecute for murder where the outcome of the assault trial was an acquittal.

    2.43         A majority of respondents agreed with our preliminary view[37] that a prosecution for murder or culpable homicide should be barred by an acquittal of the assault which was alleged to have caused the victim's death. Others disagreed, for various reasons. It was suggested that if reprosecution were only possible where there had been a conviction at the first trial, the rule would no longer be based on principle and would be difficult to justify. It was also pointed out that while common sense might suggest that an acquittal for assault was incompatible with conviction for murder arising out of that assault,[38] there were circumstances in which these verdicts could be compatible: an acquittal of assault might be secured not on the basis that the accused did not commit the actus reus, but on the basis of a defence such as compulsion or necessity which might not be available as a defence to a charge of murder.

    2.44         Each of these objections deserves to be taken seriously. The first objection is that allowing a further prosecution for homicide only where the earlier assault trial ended in conviction would not be based on principle, and so would be hard to justify. We think that the straightforward response to this is to note that the existing rule developed in the context of a rule against double jeopardy which focused quite narrowly upon preventing successive prosecution for the same crime. On Lord Ardmillan's reasoning,[39] allowing the prosecution for murder was not an exception to the rule against double jeopardy. It was rather a prosecution for a separate crime. The rule against double jeopardy did not apply because the accused had never been in jeopardy of a conviction for murder. But our proposed broader rule against multiple prosecutions arising out of the same acts is quite different. It applies to all prosecutions arising from the same acts, not merely those relating to the same charges. Principle does not compel us to recognise any exceptions. Any further prosecution arising out of the same acts - whether following an acquittal or a conviction - represents an exception to the rule which must be justified in its own terms.

    2.45         As for the other criticism, it is correct to point out that acquittal of assault and conviction of murder need not always be incompatible verdicts. The acquittal of assault may have been for a reason other than the finder of fact's having concluded that the accused did not commit the actus reus of assault. Where the acquittal was on the basis of a defence which would not be a defence to a charge of murder, there is no logical incompatibility between an acquittal of assault and a subsequent conviction of murder or culpable homicide. There are, however, significant difficulties in establishing the basis of an acquittal, particularly where the acquittal is delivered by a jury. A rule which would prevent a trial for murder where the first trial found that the accused had not committed the actus reus of assault but would allow such a trial where the acquittal had been delivered upon another basis would be quite impracticable. The choice is between an exception which would always allow a murder prosecution and one which would allow such a prosecution only following a conviction of assault.

    2.46         COPFS presented a more practical objection to limiting homicide prosecutions to cases in which the first prosecution had resulted in conviction. They noted that death changes not only the nature of the offence, but also the nature of the investigation. What seems like a straightforward assault may be the subject of only limited investigation, whereas a full murder inquiry might well produce substantial criminative evidence. Or it may have been decided to proceed with the initial assault trial without some of the available evidence, perhaps because of illness or absence of witnesses, given the relatively minor nature of the offence.

    2.47         We appreciate that, in the nature of things, fewer resources will be put into the prosecution of a common assault than into a murder. We also accept that a more thorough investigation may throw up new evidence which might have led to a different verdict at the original trial. But this is a more general point, and not restricted to homicide: all other things being equal, the more resources are devoted to investigating and prosecuting a crime, the more of the existing criminative evidence will be available. In any event we assume that, however routine the assault, the prosecutor will not take the case to court unless he has sufficient evidence to justify a conviction. On this point, the COPFS argument is one in favour of an exception on the basis of new evidence, rather than a specific exception for homicide following assault. And it would be an exception based upon evidence which, ex hypothesi, could with ordinary diligence have been made available at the original proceedings. COPFS themselves accepted that any new evidence exception should not extend to evidence which could with ordinary diligence have been available at the original trial.

    2.48         None of these criticisms is sufficient to make us alter our preliminary view, supported by the majority of respondents, that the rule against double jeopardy should prevent the prosecution of a person for murder or culpable homicide where that person has previously been acquitted of an offence involving the assault which is alleged to have resulted in the death of the victim. Accordingly we recommend:

    5. It should continue to be possible to prosecute a person for murder or culpable homicide where that person has previously been tried and convicted, prior to the victim's death, for an offence involving the assault which is alleged to have led to the victim's death.
    (Draft Bill, section 3(4))
    6. It should, however, no longer be possible to prosecute a person for murder or culpable homicide where that person has previously been acquitted of an offence involving the assault which is alleged to have led to the victim's death.
    (Draft Bill, section 3(3))
    Unintentionally causing death

    2.49         If there is to be an exception to the rule against multiple prosecution of the same acts for murder and culpable homicide following assault, should there also be an exception for other offences of causing death: for culpable homicide not involving assault, or for statutory offences of causing death (for example, causing death by dangerous driving)?

    2.50         In the Discussion Paper, we noted that such cases raise many of the same issues as a prosecution for murder. In each case, the second prosecution is in relation to a crime which could not have been prosecuted together with the earlier charge, not because of some failure or oversight on the part of the prosecution but because one of the essential elements of the offence - the death of the victim - was then absent. As with a murder in which the victim dies only after the accused has been tried for assault, the fact of death is not merely an aggravation but creates a new crime.[40]

    2.51         The Judges of the High Court of Justiciary opposed the extension of the exception to allow the prosecution of offences of causing death which do not disclose evil intent.[41] The presence or absence of evil intent does not, however, mark a sharp dividing line between serious and less serious cases. If an apparently minor assault which nevertheless leads to death may be prosecuted as culpable homicide, it would seem anomalous if a serious breach of health and safety which led to the death of one or more people could not. Indeed, in some cases, statutory offences involving no element of malice - such as health and safety and driving offences - may be more likely to lead to death or serious injury than might an assault. Furthermore, the conduct of the prosecution in choosing to prosecute prior to the death of the victim is even more clearly appropriate. While an argument might be made (however unpersuasively) that the prosecution of an assault should be delayed until it is clear that the victim will not die, the prosecution of many statutory offences is governed by strict time limits. Section 136 of the Criminal Procedure (Scotland) Act 1995 places a general six month limit on the prosecution of statutory offences which may be tried only summarily, unless the enactment in question specifies a different time limit. Among those offences triable only summarily are a number which are quite capable of having fatal consequences, the most obvious of these perhaps being careless driving.[42] It is readily foreseeable that the effect of such time limits, taken together with a general rule against successive prosecution for the same acts, could prevent certain serious crimes from ever being prosecuted.

    2.52         As an alternative to a blanket exception for culpable homicide and statutory offences of causing death, COPFS suggested that we consider the New Zealand model, where a first prosecution will bar a subsequent prosecution for homicide only if the maximum possible sentence at the original trial would have been imprisonment for a term of three years or more. Such a model would go some way to addressing COPFS's concern that a bar to a subsequent homicide prosecution should not be created by an earlier prosecution in relation to which police and prosecutors had no reason to conduct an intensive investigation. As we noted in the Discussion Paper,[43] however, there is a serious practical objection to the operation of such a system in Scotland. While New Zealand has a set sentencing tariff for all offences, here the only restriction upon the potential sentence for a common law offence is the sentencing power of the court before which the charge is brought. One could not, in Scotland, have an exception which applied uniformly to common law and statutory offences. This in itself may not be a fatal objection, since the proposal arises principally in the context of statutory offences.[44] But even if one were to limit the exception to statutory offences, it is not clear where the threshold of seriousness of the first offence should be set. There is a qualitative gap between the seriousness of an offence which involves causing death and one which does not. This is reflected in the available sentences for statutory offences: while the offence of dangerous driving carries a maximum sentence of two years in prison, the maximum sentence for causing death by dangerous driving is 14 years.[45] Should a threshold of seriousness of the first offence be set above the maximum available sentence for dangerous driving? If so, it is hard to think which statutory offences would ever bar a subsequent prosecution for an offence of causing death. But if the threshold were set below the two years available for dangerous driving, a conviction for dangerous driving would bar a subsequent prosecution for causing death by dangerous driving where the victim dies after the first trial. We do not think that this would be appropriate.

    2.53         We conclude that the same rule should apply to culpable homicide following an offence not involving an assault and to statutory offences of causing death as applies to murder and culpable homicide following assault, and recommend:

    7. It should be possible to prosecute a person for culpable homicide or for a statutory offence of causing death where that person has previously been tried and convicted, prior to the victim's death, for an offence relating to the act or omission which is alleged to have led to the victim's death; such a prosecution should, however, be barred where the outcome of the first trial was acquittal.
    (Draft Bill, section 3(3), (4))
    The problem of overlapping and inconsistent verdicts

    2.54         One of the virtues of the rule against double jeopardy is that it protects the judicial system from the risk of producing inconsistent verdicts. The recognition of any exception to the principle that the same acts should not be tried more than once introduces the possibility of inconsistent or overlapping verdicts. Verdicts may overlap where the accused is convicted of an offence which incorporates some of the conduct in relation to which he was previously tried and convicted: an obvious example might be where a person convicted of attempted murder was then prosecuted for, and convicted of, murder following the death of the victim. In such a case, that person's criminal record will misrepresent the extent of his offending, since it will appear on its face that he committed two offences (attempted murder and murder) rather than one. This might lead to the straightforward suggestion that the conviction at any subsequent homicide trial (be it for murder, culpable homicide, or a statutory offence of causing death) should be substituted for the original conviction. There will be cases, however, in which such a substitution would be inappropriate. For instance, if the original conviction were for rape or sexual assault, the substitution of a conviction for murder would produce a criminal record which was misleading in the opposite sense, since the murder conviction would not immediately disclose the sexual element of the offence.

    2.55         Of equal concern is the possibility of an incompatible verdict. Consider a case in which Agnes has assaulted Valerie, apparently with intent to kill her, and been convicted of attempted murder. After the conviction, Valerie dies. Agnes is charged with murder. One might reasonably expect Agnes to plead guilty; but what if she does not, stands her trial, and is acquitted? Perhaps her original defence was one of alibi. The first jury must have rejected this defence in order to have convicted her of attempted murder; but there seems nothing to stop a second jury from accepting the defence at Agnes's trial for murder. Or perhaps Agnes no longer insists on the alibi defence and instead says that she acted in self-defence. If Agnes is acquitted, can the attempted murder conviction stand? The answer to this question will depend crucially upon the facts of the case.

    2.56         How could inconsistent verdicts be avoided? One possibility would be to provide for the extract conviction in the first trial to be conclusive evidence at the subsequent trial of the facts set out in the libel upon which the conviction was returned.[46] This would certainly avoid the risk of inconsistency, though at some cost to the perceived fairness of the second trial. There is something uncomfortable about the prospect of the Crown's having effectively a free shot at the murder prosecution - if that prosecution results in conviction, the accused is convicted of murder; if in acquittal, the accused can go straight back to serving the remainder of his conviction for the attempt.

    2.57         An alternative approach, which would avoid giving the Crown a risk-free option to upgrade the existing conviction to one of homicide, would be to allow a prosecution for homicide only if the original conviction (let us say for assault) were first set aside. However, there would be a number of difficulties with any such proposal.

    2.58         First, it must be recognised that a prosecution may fail for any number of reasons unrelated to the guilt or innocence of the accused. Witnesses may prove unavailable, or may fail to give the expected evidence. Procedural mishaps may occur. If the original conviction were set aside before authority for a murder prosecution was granted, there would be a risk that the accused would escape conviction and punishment for the crime of which he was originally and properly convicted merely because of some mishap in the prosecution of the murder charge. The additional fact of death does nothing to cast doubt upon the original conviction, and there seems no reason why the convicted person should be afforded the windfall benefit of having a possible acquittal at the homicide trial, on purely technical grounds, result in his escaping all punishment for an offence of which he was duly convicted.

    2.59         Second, there are cases in which the first conviction and a subsequent conviction for murder could and should sit alongside one another. An example would be where the original conviction was for rape: the rape and the murder, although arising from the same acts, are clearly separate wrongs, each deserving of punishment, and it would not be appropriate to substitute the murder for the rape conviction. Nor would it be appropriate to create an incentive for the convicted rapist and accused murderer to insist on being tried again on the charge of rape as well as for the murder.

    2.60         For these reasons, we do not think that there should be a requirement that the original conviction be set aside before a homicide prosecution may be pursued, or that the result of a homicide conviction should always be the substitution of that conviction for the original conviction. Rather, it should be possible for the accused, following the second trial, to make a motion that the original conviction be quashed in light of the verdict of that trial. So where the accused is ultimately convicted of murder, having previously been convicted of attempted murder, he may make a motion that the attempted murder conviction be set aside. Or if the accused is ultimately acquitted, and the judge considers that the acquittal is incompatible with his earlier conviction, that earlier conviction should be set aside. Either party should be able to appeal against the trial judge's ruling on such a motion.

    2.61         We recommend:

    8. After the verdict at any trial for homicide following an earlier conviction for an offence relating to the act which is alleged to have caused the victim's death, the accused should be able to make a motion that the earlier verdict be quashed in light of the verdict at the homicide trial.
    (Draft Bill, section 3(4)(b))
    Verdict or sentence?

    2.62         As we noted in Part 3 of the Discussion Paper, there is presently some uncertainty about whether a guilty verdict, or the acceptance of a guilty plea, must be followed by sentence before a second trial will be barred. The authorities as they stand suggest that in summary proceedings, a plea of res judicata may be based upon the acceptance of a guilty plea by the prosecutor, regardless of whether any sentence or penalty has been imposed. In solemn proceedings, such a plea will fall to be sustained where the trial has been concluded by a determination by the jury of guilt or innocence.[47] However, the case of Pattinson v Stevenson[48] suggests that the acceptance by the prosecutor of a guilty plea in solemn proceedings before the jury is empanelled will not bar a subsequent trial. We suggested[49] that the rule should be the same in solemn and summary proceedings, and that a second prosecution should be barred wherever a plea of guilty has been accepted by the prosecution, or a verdict of guilty pronounced by the jury, regardless of whether a sentence was imposed.

    2.63         With the exception of COPFS, all respondents on this point - including the Judges of the High Court of Justiciary, the Faculty of Advocates and the Criminal Law Committee of the Law Society of Scotland - agreed with our preliminary view that a sentence should not be required in order for the rule against double jeopardy to apply. COPFS based their objection on the fact that avoiding double punishment is one of the rationales for protection against double jeopardy. We agree that this is an important aspect of double jeopardy protection, but, for the reasons explained in Part 2 of the Discussion Paper, we consider that the protection of the accused from multiple proceedings has a value independent of any concern over multiple punishment. Given this, we consider that it would be preferable to achieve consistency by increasing the scope of the protection against double jeopardy in solemn proceedings rather than by reducing the scope of that protection in summary proceedings. We recommend:

    9. The rule against double jeopardy should apply to bar a subsequent prosecution where there has been a finding of guilt, or the acceptance by the prosecutor of a guilty plea, regardless of whether sentence has been imposed.
    (Draft Bill, section 1(4))
    The treatment of foreign verdicts

    2.64         Another area in which the present law lacks clarity is in its treatment of foreign trials. Does a trial in a foreign jurisdiction bar a subsequent trial in Scotland for the same offence, or for another offence arising out of the same acts? It seems clear that the verdicts of other UK courts will be treated as barring proceedings in Scotland;[50] the Schengen convention prevents the prosecution in the UK, on the basis of the same acts, of those who have previously been tried in other EU states;[51] but the status of other foreign verdicts is unclear.

    2.65         In the Discussion Paper, we suggested that if the rule against double jeopardy is justified by the accused person's interest in finality, and by the desire to avoid repeatedly subjecting individuals to the stress of the criminal process, then it should not matter where the original proceedings took place. We recognised, however, that there would be cases in which it would be inappropriate to treat foreign verdicts as barring proceedings in Scotland. In some cases, there would be grounds for believing that a purported trial was aimed at shielding the accused from responsibility for his crimes rather than at calling him to account for them. In others, there might simply be too great a divergence between the foreign legal system and our own, whether in relation to procedure, substantive law, or the seriousness with which particular conduct is viewed.[52]

    2.66         We suggested that the default position should be that foreign judgments should be respected, but that a court should have the discretion to disregard such judgments where appropriate. In deciding when it would be appropriate to disregard a foreign verdict, we suggested that the court should have regard to a number of considerations modelled on the precedent of Article 20(3) of the Rome Statute on the International Criminal Court. These were whether it appeared that the foreign proceedings: (a) were held for the purpose of shielding the accused from criminal responsibility; (b) were not conducted independently or impartially; or (c) were conducted in a manner which was inconsistent with an intent to bring the perpetrator of the alleged offence to justice.

    2.67         There was almost universal support for the view that foreign judgments should in principle be respected for the purposes of double jeopardy protection. The only outright dissent came from Professor Paul Roberts, who noted that double jeopardy protection is usually sovereign-specific, criminal jurisdiction is usually territorially based, and that foreign judgments in criminal cases do not receive automatic recognition in English courts but are treated as questions of fact. He suggested that where a single act or course of conduct simultaneously constituted an offence in two legal systems, there was a real sense in which two separate wrongs had been committed against two discrete legal systems, each of which might legitimately seek redress.

    2.68         We think that each of these criticisms can be answered. Our proposal does not call for the automatic recognition of foreign verdicts as such. The foreign verdict will not have any direct effect in Scots law; rather, it will be for the accused to establish the existence and significance of the previous proceedings as a plea in bar of trial, and the court will retain a discretion to disregard the foreign proceedings if this is in the interests of justice. In that connection we note the recent judgment of the Court of Final Appeal of Hong Kong, in which Bokhary PJ said -

    "There is a discretionary power to stay a prosecution as an abuse of process where (i) a person faces a second trial arising from the same or substantially the same set of facts as gave rise to an earlier trial (whether in the same jurisdiction or in a competent court in another jurisdiction) and (ii) the prosecutor cannot advance any special or exceptional circumstances to justify the holding of a further trial."[53] (emphasis added)

    2.69         Whether or not one accepts the view that the existence of two distinct wrongs against two distinct sovereigns should justify successive proceedings in respect of the same act[54] depends upon the view that one takes of the motivating rationales of double jeopardy protection.

    2.70         We place considerably more weight than does Professor Roberts upon the value of protecting the accused from the stress of multiple proceedings, and consequently find the idea of successive prosecution in respect of the same acts to be at least prima facie objectionable, regardless of whether the two prosecutions relate to distinct legal wrongs.

    2.71         Both Professor Roberts and James Chalmers queried our proposed borrowing from the Rome Statute of the International Criminal Court, pointing out that the ICC, being possessed of an international jurisdiction, was in a very different position to a domestic court whose jurisdiction was (almost) exclusively territorial. Professor Roberts suggested that for a Scottish court to hold that foreign proceedings were held for the purpose of shielding the accused from criminal responsibility, or that they were not conducted independently or impartially, would risk diplomatic embarrassment and potentially serious consequences for international relations. The point is well made; however, if a Scottish court were satisfied that a foreign prosecution had been a sham, aimed only at shielding the accused from criminal responsibility, it is hard to see why the interests of justice should be subordinated to a desire to avoid diplomatic embarrassment. Mr Chalmers drew our attention to the provisions of Article 2(1) of the Convention Between the Member States of the European Communities on Double Jeopardy, which permitted signatories to declare that they would not be bound by the proposed prohibition on cross-jurisdictional double jeopardy, inter alia:

    "(a) If the facts which were the subject of the judgment took place on [the Member State's] own territory either in whole or in part. In the latter case this exception shall not apply if those facts took place partly on the territory of the Member State where the judgment was rendered."
    He suggested that a similar consideration should apply in relation to the recognition in Scotland of foreign verdicts: if a crime is committed in Scotland it should not be possible for the courts of another country to deprive the Scottish courts of jurisdiction.

    2.72         COPFS expressed cautious support for our proposed approach, but pointed out a number of difficulties:

    "It is not just a question of bad faith as the considerations suggested infer: there may be cases where the issue is around poor quality of investigation or prosecution, or where the authorities of the other country have acted properly within their own system and culture, but either they or their system does not deal with the acts with anything approaching the level of seriousness applied in this country. An example might be if a country's laws equiparated drug smuggling with tobacco smuggling and provided for very low penalties. Given that the UK generally proceeds on the basis of territorial jurisdiction such cases are more likely to arise where the foreign jurisdiction recognises and is applying extraterritoriality. An approach which recognised further considerations or which created a general presumption against any bar, subject to a plea of oppression, might be a safer approach."

    2.73         We have already rejected a general presumption in favour of disregarding foreign proceedings as incompatible with our view of the purpose of the rule against double jeopardy as protecting the accused from the stress of multiple proceedings. Nor do we think that it would be appropriate to fall back upon the existing plea of oppression, for the reasons given above at paragraph 2.25. We do consider, however, that a court should be able to refer to factors such as territoriality and the relative seriousness with which the foreign jurisdiction treats the conduct forming the subject matter of a charge in deciding whether or not it would be in the interests of justice to allow a prosecution to continue notwithstanding the existence of a prior foreign verdict in relation to the same acts.

    2.74         We therefore recommend:

    10. A verdict of conviction or acquittal by a foreign court should bar a subsequent prosecution in Scotland in respect of the same acts which gave rise to the first prosecution; but the Scottish court should be permitted to disregard an acquittal or conviction in a foreign jurisdiction where it is satisfied that it is in the interests of justice to do so.
    (Draft Bill, section 2(4), (5)(a))
    11. In considering whether to disregard a foreign verdict, the court should have regard, among other considerations, to –
    (a) the extent to which the acts constituting the offence charged are alleged to have taken place in Scotland;
    (b) the relative seriousness with which the law of Scotland and that of the foreign jurisdiction view the conduct which forms the subject matter of the instant charge;
    (c) whether it appears that the foreign proceedings were held for the purpose of shielding the accused from criminal responsibility, were not conducted independently or impartially, or were conducted in a manner inconsistent with an intent to bring the perpetrator of the alleged offence to justice.
    (Draft Bill, section 2(6))
    The incidental proof of foreign offences

    2.75         While we recommend that foreign acquittals or convictions may bar subsequent proceedings in Scotland in respect of the same acts, this should not be interpreted as preventing the prosecution from proving the commission of offences abroad as part of the evidential background to a Scottish charge. The classic example of such a case is Dumoulin v HMA[55] in which evidence of an alleged fraud in Germany was led as part of the Crown case that Dumoulin had murdered his wife by pushing her over a cliff. Such evidence was properly led to show that Dumoulin had a motive for the crime, having arranged his affairs, prior to his marrying the deceased, in such a way that he would benefit from her death.[56] We do not consider that our proposal on the recognition of foreign verdicts should prevent the leading of such evidence of foreign offences in the future, even if the accused has already been tried in relation to the foreign conduct which is alleged to form the background to, or a minor part of, the Scottish offence.

    Nullity of first proceedings

    2.76         One of the established features of the present Scots law of double jeopardy is that, in order for the rule against double jeopardy to apply, the first proceedings cannot have been fundamentally null (so, for instance, there is no bar to a further trial where the original court was improperly constituted, or the original charge failed to specify the locus of the offence).[57] This is in accordance with principle - double jeopardy cannot arise where the accused was never truly in jeopardy at the first trial - and with the rules of all of the other legal systems which we surveyed.[58]

    2.77         Application of the letter of the law could, however, result in an accused who has endured the stress of one set of proceedings which he believed to be valid being faced with a second set of proceedings in respect of precisely the same offence. As a matter of principle this is acceptable; indeed, the principle that every criminal charge should be determined by a fair and proper trial will generally demand that a valid trial be held, even if this means subjecting the accused to the stress of repeated proceedings. We nevertheless suggested in the Discussion Paper that it might be appropriate for the court to be able to bar such proceedings, in particular cases, where the defect in the original trial is not attributable to the fault of the accused and such further proceedings would place him under unnecessary strain.

    2.78         There was widespread agreement with this proposal. One respondent was opposed, on the basis that if there were such a power it would be difficult to know when it had been, or could be, properly exercised. Another took the opposite view, supporting the proposed discretion but arguing that it would be better dealt with as part of a more general discretion to halt proceedings which would constitute an abuse of process.

    2.79         COPFS were opposed to any such discretion, suggesting that there would be a risk that the accused might choose not to raise any issue of fundamental nullity in order to rely on it later to avoid prosecution. They suggested that adequate protection was already afforded by the existing rules relating to oppression and timebar.

    2.80         We note that none of the modern cases to which we referred in the Discussion Paper concerned an attempt to prosecute again after the original proceedings had concluded in an ex facie valid verdict. In Dunlop v HMA,[59] no plea of res judicata was available as the verdict which followed the acceptance by the Crown of the accused's guilty pleas was delivered by only 14 members of the original jury. Such a case would now be dealt with by our recommendation that the acceptance of a guilty plea by the prosecutor should be sufficient to trigger the rule against double jeopardy. In Whitelaw v Dickinson[60] the first proceedings, being time-barred, did not conclude in a verdict. Although cited as a case of double jeopardy, Whitelaw turned not upon the common law rule of tholed assize but upon the interpretation of section 101(1) of the Criminal Procedure (Scotland) Act 1975.[61] It did not concern proceedings which had resulted in an ex facie valid verdict which was later found to be a nullity. Similarly, in both Thomson, petitioner[62] - in which the original indictment failed to specify the locus of the alleged offence - and McGlynn v HMA[63] - where the original indictment referred, incompetently, to a statutory offence which was triable only on complaint - the defect was discovered in the course of the original proceedings. Alison's report of the case of William McLellan[64] does not tell us whether the presence of a minor on the jury was discovered during or after the original proceedings.

    2.81         We consider that the risk of real unfairness to the accused would arise only in the case where the defect in the original proceedings was asserted by the Crown at some point after those proceedings had terminated in an ex facie valid verdict. We are aware of no such case having arisen in practice. We expect that it would be in only the most exceptional case that a court would consider disregarding an ex facie valid verdict which had not been set aside by a higher court.[65]

    2.82         We consider that the Crown's concern about the risk of the accused deliberately failing to raise an issue of fundamental nullity could appropriately be dealt with by the court in considering whether to exercise its discretion. We doubt if a court faced with evidence that the accused had deliberately sought to manipulate its process in this way would be prepared to exercise the discretion. Nor do we think that it would be appropriate simply to rely upon a plea of oppression. In order for such a plea to succeed, it seems that the court must be satisfied that the risk of prejudice is such that the accused could not reasonably be expected to have a fair trial.[66] The proposed discretion is of much the same character as a plea of oppression, but we do not think that the existing rules of oppression and timebar adequately address the case where the accused has, through no fault of his own, been subjected to invalid proceedings. In such a case, at least where the accused has not been at fault, it may be inappropriate to bring further proceedings regardless of whether those proceedings, viewed in isolation, might be thought fair. In any case, if the law in this area is to be clarified and codified, to fall back upon the common law plea of oppression - the requirements of which are poorly understood - would be to leave a lacuna.

    2.83         On the basic question, while the situation is likely to arise only rarely, we consider that any attempt by the prosecutor to prosecute again where it is contended that the first proceedings were fundamentally null should be subject to the control of the High Court. That Court should have the power to prevent such a prosecution - even if it is satisfied that the first proceedings were a nullity - in circumstances where to proceed against the accused would be contrary to the interests of justice.

    2.84         We recommend:

    12. In any case where the Crown would propose to argue in response to a plea which might be raised in relation to double jeopardy that the original proceedings were a nullity, the approval of the High Court should be required before proceedings may be brought.
    (Draft Bill, section 10)
    Part 3 Tainted Acquittals
  5. 3.1         As we observed in the Discussion Paper,[67] there may be cases in which the trial process is subverted or perverted by someone bribing or threatening witnesses, jurors or, in extreme cases, the judge. Such an interference with the trial process raises the question of whether there has been a valid trial at all. Since the whole basis of the plea of res judicata rests upon a proper decision reached in accordance with the relevant rules, we argued that there can be no public interest in recognising finality in a judgment produced otherwise than in accordance with those rules. Given this, we asked whether it would be appropriate to recognise an exception to the rule against double jeopardy in such cases. On reflection, we think that we were wrong to speak of an exception. Allowing for the retrial of those whose acquittals are tainted by interference with the trial is not an exception to the rule, but a vindication of it.

    3.2         We cited perhaps the most striking reported case on the topic, Aleman v Honourable Judges of Cook County,[68] in which the original acquittal had followed the accused's bribery of the judge. When the bribery became apparent, the original accused was re-indicted on the murder charge and pled the 5th Amendment bar on double jeopardy, pointing to a series of decisions in which the Supreme Court had held that an acquittal on a charge unequivocally bars retrial on that charge. The United States Court of Appeals for the Seventh Circuit, reviewing the decision of the Illinois courts to allow a fresh prosecution for murder, said:

    "The Illinois courts viewed the authority cited by Aleman as begging the question; the Double Jeopardy rule may well be absolute when it applies . . . but determining if it applies is the real issue in this case. Similarly, the State argues that the protections of the Double Jeopardy Clause only extend to a defendant who was once in jeopardy of conviction on a particular criminal charge; the State contends that, by bribing Judge Wilson, Aleman created a situation in which he was never in jeopardy at his first trial. The first trial, therefore, was a sham and the acquittal there rendered has no effect for double jeopardy purposes. Under this theory, the state was free to re-indict him because he has never been in jeopardy of conviction on the Logan murder charge."[69]
    The Court went on to hold that:
    "To allow Aleman to profit from his bribery and escape all punishment for the . . . murder would be a perversion of justice, as well as establish an unseemly and dangerous incentive for criminal defendants . . . For these reasons, we affirm the district court's rejection of the double jeopardy claims contained in Aleman's petition."[70]

    3.3         The case of Aleman neatly illustrates the two strands of reasoning which may justify a new trial where the original acquittal was tainted. The first approach focuses upon the validity of the procedure, arguing that the effect of the interference was to prevent there having been a proper trial: there is no double jeopardy, since the accused was not truly in jeopardy at the first trial. The second approach stresses considerations of equity and public policy: it would be wrong to allow a person who had committed an offence against the administration of justice to benefit from this conduct.

    3.4         We asked whether it should be possible for a person to be tried again where that person's previous acquittal was tainted by an offence in relation to the administration of justice[71] and, if so, whether it should be possible to have such a retrial where it is established that the original accused played no part in the administration of justice offence.[72]

    3.5         Without exception, those who responded to the Discussion Paper on this point were in favour of allowing retrials following a tainted acquittal. Respondents were almost evenly divided on the question of whether a retrial should be possible where it could be shown that the accused had nothing to do with the tainting of his first trial. Some took the view that the crucial question was whether the first trial was in fact tainted, so as not to be a valid trial.[73] On this view, there having been no valid first trial, it was irrelevant whether the accused had been involved in the tainting offence. Others regarded the accused's involvement as essential to the fairness of allowing a retrial. The Criminal Law Committee of the Law Society of Scotland commented that, "it would be unfair to try an accused person twice if he had not been responsible for, or in some way participated in, the criminal act which led to the perversion of the original trial process."

    3.6         On balance, we incline to the view that it should not be necessary to show the involvement of the accused in the tainting of the original trial. There are practical reasons for this: it would, in practice, often be difficult to prove a connection between the accused and the person who bribes or intimidates witnesses or jurors. More important, however, is the principled argument that what justifies the second trial is the fact of the perversion of the first: if there was no proper first trial, then there should be no legal bar to a second trial of the same charge. There is a strong analogy with the case (discussed above at paragraphs 2.76 to 2.81) in which the first trial is discovered to have been a fundamental nullity. There is clearly merit in the equitable argument that the accused who acts to pervert his first trial should not be permitted to benefit from this wrongdoing; where it was the accused who was responsible for the interference in the original trial, then he can have no claim that it is unfair to try him again. But the true - and sufficient - justification for the second trial is the defective nature of the original proceedings, not the desire to punish the accused for his involvement in the tainting offence.

    3.7         One of the respondents to the Discussion Paper supported allowing for retrial following a tainted acquittal but cautioned against describing this as an exception to the rule against double jeopardy. Rather than representing an exception, the tainted acquittal procedure could be seen as a vindication of the core principle underlying the rule against double jeopardy, namely that every criminal charge should be the subject of one proper trial.[74] We agree. Viewed as a question of whether there has been a proper trial at all, allowing a second trial following at tainted acquittal can be seen as a coherent part of the rule against double jeopardy rather than an exception to it.

    3.8         We recommend:

    13. It should be possible to retry an acquitted person where that person's acquittal was tainted by an offence against the administration of justice in relation to the original trial.
    (Draft Bill, section 4)
    14. It should not be necessary, in order to justify such a retrial, to show any involvement of the acquitted person in the offence against the administration of justice.
    (Draft Bill, section 4(3)(a))
    Perjury as a taint

    3.9         One question which we did not address in any detail in the Discussion Paper is which offences should be regarded as having a tainting effect on the original proceedings; more specifically, should simple perjury count as a tainting offence?

    3.10         We are firmly of the view that it should not. It is in the nature of the judicial process that much of the evidence will be inaccurate, owing in many cases to failures in recollection by witnesses and, in some cases, to a desire on the part of witnesses to hide the truth or suggest what is not true. In any case where there is a conflict of testimony, one or other - or all - of the witnesses must be either honestly mistaken or lying. No doubt in some cases there is perjury. It is the function of the fact finder - the jury or judge - to balance conflicts of testimony, taking into account their impressions of the witnesses, and to arrive at a view of what actually happened. Since, at least in the case of the jury, their verdict does not disclose how they arrived at their decision, it is normally very difficult to say on what basis they appear to have accepted one piece of evidence and rejected another. The assessment of whether a witness is inaccurate, evasive, or actually guilty of perjury, is part of the normal trial process, in a way that external interference - the subornation of perjury - is not. To allow the outcome of a trial to be revisited in the light of further investigations into the truthfulness of the testimony of witnesses would be significantly to undermine the finality of the typical criminal verdict.

    3.11         What about perjury by the accused? Here the answer is more complicated, since the equitable considerations identified above at paragraph 3.3 apply with full force: if the accused has sought, by dishonest means, to alter the course of his trial, then he should not, in justice, be allowed to benefit from this misconduct. In the Discussion Paper we noted that where an accused person secured his acquittal by giving false evidence, it was possible to prosecute him for perjury where evidence subsequently came to light which demonstrated the falsity of his testimony at the original trial.[75] We suggested that it should continue to be possible to prosecute the original accused for perjury in such circumstances.[76] This proposal attracted widespread support from the respondents to the Discussion Paper.

    3.12         We accordingly recommend:

    15. It should continue to be competent to prosecute a person for perjury where there is evidence that he committed perjury in giving evidence on his own behalf in prior criminal proceedings.

    3.13         We then considered whether anything further should happen in a case similar to HMA v Cairns,[77] in which the original accused, having given evidence at his trial denying his guilt, and having been acquitted, subsequently confesses. While it is appropriate that such a person should be tried and punished for perjury, conviction of perjury does not reflect the full extent of his wrongdoing. We suggested that it should be possible, in such circumstances, for the accused to be tried again. There was broad support among respondents to the Discussion Paper for this position.

    3.14         Any proceedings based upon the perjury of the accused at the original trial will necessarily be based upon new evidence suggesting that the accused's sworn denial of guilt was a lie; that is, new evidence showing the accused to be guilty of the crime with which he was charged. If there is such new evidence, it seems to us that it is on the basis of this new evidence, rather than on the basis of the accused's perjury, that the justification for any new prosecution must rest. Furthermore, where the new evidence takes the form of a post-acquittal admission of guilt, we do not think that there is any meaningful distinction to be drawn between cases in which the accused did or did not give evidence on his own behalf at the original trial.[78] For these reasons, we deal with post-acquittal confessions in the context of new evidence, in Part 4 below, rather than as examples of tainted acquittals.

    The test of a tainted acquittal

    3.15         In the Discussion Paper, we asked a number of questions concerning the procedure by which a new trial might be authorised following a tainted acquittal. We asked whether a retrial should require to be authorised by a court following the conviction of some person for an offence against the administration of justice in relation to the original trial;[79] whether the process leading to the authorisation should include, in addition to the conviction of someone for such an offence, a judgment by the court that the offence had a significant effect on the result of the previous trial;[80] and, if so, what test the court should be required to apply in assessing that effect.[81] We also suggested that the court should have an overriding discretion to refuse to authorise a retrial where such a refusal appears to the court to be in the interests of justice,[82] and that it should be possible to impose reporting restrictions upon the application proceedings with a view to preventing publication of their result from influencing the jury in any subsequent trial.[83]

    3.16         With the exception of COPFS, whose response is considered in some depth below, all those who responded on these points agreed that the authorisation of a new trial on the basis of a tainted acquittal should require that some person - not necessarily the original accused - should have been convicted of an offence against the administration of justice in relation to the original trial. There was general agreement that the prosecution should also be required to show that the offence had, or was likely to have had, a significant effect upon the original trial, though opinions differed as to what the appropriate test should be. It was also generally agreed that the court should be able to refuse to authorise a retrial where a refusal appears to the court to be in the interests of justice, and that it should be possible to impose reporting restrictions to reduce the risk of prejudice to any subsequent trial.

    The purpose of having a test

    3.17         If the principal justification for allowing the retrial of a person whose acquittal has been tainted by interference with the course of justice is that there was no proper first trial - that that person was never truly in jeopardy - then it would be quite logical to suggest that no further procedure would be needed before the accused could be re-indicted. This appears to have been what happened in Aleman: in the light of the overwhelming evidence that the original acquittal had been secured by the bribery of the trial judge (who had, shortly after receiving a grand jury citation in respect of alleged corruption, committed suicide), Aleman was simply served with a new indictment for murder. As the State argued before the Court of Appeals, the first trial had been a sham and the acquittal was of no effect for double jeopardy purposes.

    3.18         There is a strong analogy with the case in which the first proceedings were a fundamental nullity, as where the case was held before an improperly constituted court, or the verdict delivered upon an incompetent charge.[84] In such cases there is presently no procedure which must be gone through before new proceedings may be brought against the original accused; rather, the fundamental nullity of the first proceedings would form a good answer to any plea of res judicata which might be made by the accused in relation to the second indictment. It appears that the same approach was taken in the cases cited by Hume in which the High Court dismissed pleas of res judicata on the basis that criminal proceedings held before the Justices of the Peace in Forfar and the Magistrates of Dunbar had been collusive, mock trials.[85]

    3.19         Nevertheless, we reject this approach. As we have repeatedly observed, there is considerable value in the finality of judicial verdicts. This value is respected by the rule against double jeopardy, which protects a person who has once had a verdict delivered upon a particular charge from being tried again in relation to the same matter. While we think that it would be appropriate to remove this protection in cases where the first verdict was obtained in tainted proceedings, we do not think that it would be appropriate to allow the prosecution to proceed on the basis that an acquittal was tainted without the approval of the court.

    3.20         What, then, should be the test? From the preceding discussion, it is clear that the principal aim of any such test should be to identify those cases in which there has been interference in the conduct of the original trial, such that it was not truly a proper trial. In order to reach such a conclusion, the court considering an application must be satisfied: i) that an offence against the administration of justice was committed; ii) that that offence is likely to have had an effect upon the original proceedings; and iii) that that effect is, or is likely to have been, sufficiently serious to justify the conclusion that the original trial was not a proper trial.

    The English example

    3.21         In relation to the test to be applied in assessing the effect of the offence against the administration of justice upon the course of the original trial, a number of respondents referred to the tests which apply in England and Wales by virtue of sections 54 to 57 of the Criminal Procedure and Investigations Act 1996. That Act imposes three requirements before a retrial is authorised. First, some person must be convicted of an administration of justice offence involving the bribery or intimidation of witnesses or jurors.[86] Second, the court which finds that person guilty of that offence must certify that, in its opinion, there is a "real possibility" that but for the commission of that offence the acquitted person would not have been acquitted.[87] Finally, the High Court, on an application supported by that certificate, must consider it "likely that, but for the interference or intimidation, the acquitted person would not have been acquitted."[88]

    3.22         A number of respondents were drawn to the test of "real possibility". As we shall suggest below, such a reasonably low test may have its merits; but it is important to note that the test as it applies at present in England and Wales is not ultimately that of whether there is a "real possibility" that the outcome of the original trial would have been different had it not been for the interference, but rather whether it is "likely" that "but for" the interference or intimidation, the acquitted person would not have been acquitted.

    3.23         There is reason to think that the 1996 Act sets the bar too high. As the Director of Public Prosecutions confirmed to us, in the more than a decade since sections 54 to 57 came into force, they have never been used. It is of course possible that the reason for this is that there has not, since 1996, been a case in which an acquittal was secured by the bribery or intimidation of witnesses or jurors. Alternatively, the lack of cases may be the result of a set of conditions which, cumulatively, are exceedingly difficult to meet. There are two respects in which a test such as that contained in the 1996 Act is difficult to satisfy. First, it will often be difficult to secure a conviction for an offence against the administration of justice. Second, it is asking a great deal for a court to conclude that but for some interference with witnesses or jurors, the result of the original trial would not have been acquittal.

    The need for a conviction

    3.24         COPFS emphasised the first of these difficulties. They suggested that there could be circumstances in which there was reliable and credible evidence of, for example, jury tampering without there being a sufficiency of evidence against any one individual. In such circumstances, they suggested, it may be inappropriate to require a conviction as a precondition for retrying the accused.

    3.25         This suggestion is to be taken very seriously. It is easy to imagine the difficulties attendant upon successfully prosecuting even the clearest cases of interference with witnesses or jurors. Typically, such interference will take the form of anonymous threats. At the very least, any threats or bribery will generally take place in private, making it next to impossible to secure the corroborated evidence required to make a sufficient criminal case against any individual. It may be relatively straightforward to establish, with a good degree of confidence, that interference has taken place, but almost impossible to secure a conviction. Sometimes it might actually be impossible to secure a conviction, as where the person responsible for the interference has died or disappeared. In any event, the requirement to secure an actual conviction is illogical. Even if someone is prosecuted for an offence against the administration of justice, it may be that the prosecution will fail not because of any doubt that an offence has been committed, but because it cannot be proved to have been committed by the person who is accused.

    3.26         We have already suggested that it does not matter who is responsible for the tainting of the original trial; what matters is simply whether the original trial has been tainted. It follows that it would be proper to allow a retrial where it could be shown that the original trial had in fact been tainted, regardless of whether the author of the taint could be identified, still less convicted. A procedure which required an actual conviction would have the merit of simplicity, since a conviction would unequivocally establish the fact of interference. But we have come to the conclusion that it would be appropriate to allow for reprosecution regardless of whether there has been a conviction for the tainting offence, provided that the court can be satisfied that the offence has actually been committed.

    3.27         If this is so, then the question arises as to the standard of proof which should apply in establishing the commission of the tainting offence. We consider that the commission of the tainting offence should be proved on the balance of probabilities. Two reasons inform this conclusion. The first is that the purpose of allowing tainted acquittals to be identified without the need for a conviction is to ease the otherwise formidable problem of proof which would face the prosecution. To require proof of the commission of the offence to the criminal standard would be to reintroduce the very difficulties which the non-conviction route is intended to avoid. In particular, there would be room for argument that proof to the criminal standard required the application of criminal rules of evidence, including the requirement of corroboration. If the typical case of interference with a witness is threats made in private, it is most unlikely that such corroborated evidence will be available, even to establish that the threat was made. The second reason is that, on this approach, the question of whether the tainting offence took place is a disputed question of fact, and the general rule is that disputed questions of fact - or at least those that do not themselves have penal consequences - should be determined according to the civil standard, that is on the balance of probabilities.[89]

    3.28         We recommend:

    16. Before authorising a new prosecution on the basis of a tainted acquittal, it must be established, on the balance of probabilities, that an offence against the administration of justice was committed in relation to the original trial.
    (Draft Bill, section 4(3)(a))
    17. The commission of such an offence may be proved either by the production of a relevant extract conviction, or by evidence led by the prosecutor.
    (Draft Bill, section 4(3)(a))
    The nature of the test

    3.29         The other high hurdle which must be surmounted before a new prosecution may be authorised in England and Wales is the requirement that the High Court be satisfied that but for the interference or intimidation the outcome of the original trial would not have been acquittal. We find this requirement problematic in three respects.

    3.30         First, it is unclear what the test actually means. In order to convict, a jury must be satisfied beyond a reasonable doubt. What does it mean to say that it is "likely" that, had there not been interference, the outcome of the original trial would not have been acquittal? This must be equivalent to saying that it is likely that but for the interference, the original jury would have found the accused guilty. But is it possible for a court to say that it is likely that a (hypothetically reasonable) jury would have convicted without, in substance, itself reaching - and expressing - the view that the accused is in fact guilty?

    3.31         Second, if the test does effectively call for the court hearing the application to express its own view of the accused's guilt, then it is undesirable for that reason. It would be preferable to avoid the court's having to express any such view, since questions of guilt or innocence are properly for the jury. Further, the court's decision may run the risk of prejudicing a subsequent jury.[90]

    3.32         Third, it seems to us that such a test will often be practically impossible to meet, even in circumstances where there has clearly been significant interference with the original trial. A likely scenario for the tainting of a trial would involve interference with one or more of the prosecution witnesses, resulting in the insufficiency of the Crown case and the acquittal of the accused on a successful plea of no case to answer. In such a case, there would simply be no evidential basis for assessing the likely outcome of the trial, since proceedings may have been brought to a premature end by the collapse of the Crown case, without all of the prosecution evidence being heard, and without any of the defence evidence having been led.

    3.33         What are the alternatives to the test in the 1996 Act?

    "Miscarriage of justice"

    3.34         A number of respondents suggested a test of "miscarriage of justice", building on the existing concept in appeals against conviction under section 106 of the Criminal Procedure (Scotland) Act 1995. The Faculty of Advocates suggested that the test should be whether the court is satisfied that the offence against the administration of justice led to a miscarriage of justice in the original trial, and referred to the test of miscarriage of justice set out by the High Court in Megrahi v HMA:[91]

    " . . . the next question that requires to be addressed is the content of the proposed additional evidence, and whether its significance is such that the fact that it was not heard by the trial court would be regarded as having resulted in a miscarriage of justice . . . We summarise the approach adopted in those cases in the following propositions: (1) The court may allow an appeal against conviction on any ground only if it is satisfied that there has been a miscarriage of justice. (2) In an appeal based on the existence and significance of additional evidence not heard at the trial, the court will quash the conviction if it is satisfied that the original jury, if it had heard the new evidence, would have been bound to acquit. (3) Where the court cannot be satisfied that the jury would have been bound to acquit, it may nevertheless be satisfied that a miscarriage of justice has occurred. (4) Since setting aside the verdict of a jury is no light matter, before the court can hold that there has been a miscarriage of justice it will require to be satisfied that the additional evidence is not merely relevant but also of such a significance that it will be reasonable to conclude that the verdict of the jury, reached in ignorance of its existence, must be regarded as a miscarriage of justice. (5) The decision on the significance of the additional evidence is for the appeal court, which will require to be satisfied that it is important and of such a kind and quality that it was likely that a reasonable jury properly directed would have found it of material assistance in its consideration of a critical issue at the trial. (6) The appeal court will therefore require to be persuaded that the additional evidence is (a) capable of being regarded as credible and reliable by a reasonable jury, and (b) likely to have had a material bearing on, or a material part to play in, the determination by such a jury of a critical issue at the trial."

    3.35         The court's analysis of the term "miscarriage of justice" is a comprehensive and attractive formulation of the approach to be taken in considering the likely effect of new material on a previous trial which has led to a conviction. While much of what it says is equally applicable to the approach to be taken in regard to a previous acquittal, it is not necessary, and may be misleading, to label the two processes with the same term.

    The proposed test

    3.36         We have set out various versions of what the test for a tainted acquittal should not be; but what should it be?

    3.37         It is helpful to consider the various means by which an offence against the administration of justice might have an effect upon the trial. Here, the most significant differences between offences involving the bribery or intimidation of witnesses and those involving interference with jurors (or, however unlikely, with the judge). Each of these cases calls for a different test.

    Interference with witnesses

    3.38         Interference with witnesses may involve the bribery or intimidation of a prosecution witness or the suborning of a defence witness to give false evidence.

    3.39         Where the interference is with a prosecution witness, it may have a number of consequences. It is possible that the witness will stand firm in the face of intimidation or the offer of a bribe and will give exactly the evidence that he would have given had the attempt at bribery or intimidation not taken place. Alternatively, the bribery or intimidation may succeed in inducing the prosecution witness to change his evidence, or to refuse to give evidence at all. The loss or alteration of this evidence may have one of three consequences. The evidence might have been essential to the corroboration of the Crown case. If so, the loss or alteration of that evidence as a consequence of the interference may lead directly to the acquittal of the accused on a submission of no case to answer. Alternatively, the evidence, while not essential to the corroboration of the Crown case, might nevertheless have been essential to its strength. If there is no reasonable prospect of conviction in the absence of the evidence then the Crown may well choose to abandon the case. The third possibility is that the loss or alteration of the evidence, while weakening the Crown case, is not sufficient to prevent that case from proceeding to the jury. In such a case the significance of the interference with the administration of justice, and the consequent loss or alteration of the expected Crown evidence, is rather harder to judge.

    3.40         The offence may also consist in the suborning of a defence witness to give false evidence on behalf of the accused. An obvious example would be a case in which the accused prevailed upon an acquaintance to provide him with a false alibi.[92] In the typical case of suborning perjury, the result of the offence will be the introduction of false testimony. Unlike the case of interference with Crown witnesses, the evidence introduced by the witness whose perjury was suborned cannot result in the collapse of the prosecution case as insufficient. The effect of the introduction of the false evidence will depend upon the influence which that evidence has upon the jury's deliberations.

    3.41         Broadly then, interference with witnesses may have one of three consequences. It may have no effect (or only a minimal effect) upon the evidence given by the witness, and so no effect (or only a minimal effect) upon the course of the trial. It may prevent the hearing of evidence which would be crucial to the sufficiency of the Crown case, or without which the Crown case would not be strong enough to proceed to the jury, leading to the acquittal of the accused without the defence case being heard. Or it may result in the withholding of evidence which would have had, or the introduction of false evidence which may have had, a material bearing upon the jury's consideration of a critical issue at the trial.

    3.42         We consider that it would be proper to speak of the original trial been tainted in either of the latter two cases. We recommend:

    18. An acquittal should be regarded as tainted where, but for the commission of an offence against the administration of justice involving interference with a witness, the witness in question would have either –
    (i) given evidence which was capable of being regarded as credible and reliable by a reasonable jury and which would, if accepted by the jury, have had a material bearing on, or a material part to play in, the determination by them of a critical issue at the trial; or
    (ii) refrained from giving perjured evidence which, assuming that it was accepted by the jury, was likely to have had a material bearing on, or a material part to play in, the determination by them of a critical issue at the trial.
    (Draft Bill, section 4(14))

    3.43         This test is of course borrowed from points (5) and (6) of the passage which we have cited from Megrahi.[93] For the reasons which we have already given, we think that it is appropriate to separate out this element of the test which the High Court identified, rather than simply referring to the more general concept of miscarriage of justice.

    Interference with jurors or judge

    3.44         The other type of interference with the administration of justice is interference with jurors, or indeed the judge. While we consider the prospect of the intimidation, bribery or a corruption of a Scottish judge of the modern era[94] to be quite fanciful, the possibility cannot be excluded as a matter of principle, and provision should be made even for such unlikely contingencies.

    3.45         While it is possible to set out a reasonably detailed test for assessing the effect of interference with witnesses, interference with jurors does not lend itself to such analysis. There does not seem to be any sound basis for applying a test based upon the likely outcome of trial had interference not taken place. In any case where the issue goes to the jury, there will have been sufficient evidence to entitle a reasonable jury to convict; if not, the issue would have been raised by the defence as part of a submission at the close of the prosecution case. The issue which arises in the case of interference with jurors is that the interference suggests that the evidence has not been properly considered by the jury. In one sense, interference with the jury raises a more fundamental issue than interference with witnesses, since if the tribunal is not independent and impartial there can never be a fair trial, regardless of the state of the evidence. Further, as noted above, there is little or no information as to how the jury conducts its business: only the outcome is made public.

    3.46         We would therefore suggest that the test to be applied in the case of interference with jurors should be a relatively low one, aimed simply at excluding those cases in which the court can establish that the interference is unlikely to have affected the verdict. Such cases would include the case in which the interference with the jury was brought to the attention of the trial court and the trial judge decided that it would be fair to continue with the trial,[95] and - at least as a matter of practice - those cases in which the acquittal was by unanimous verdict of the jury.

    3.47         Given the importance of the independence of the jury, we think that it would be appropriate to regard an acquittal as tainted unless the court is satisfied that the interference has had no effect on the jury's consideration of its verdict.

    3.48         We recommend:

    19. Where it is established that there has been interference with a juror or with the judge in the proceedings in which the accused was acquitted, that acquittal should be regarded as tainted unless the court hearing the application is satisfied that the interference has had no effect on the proceedings.
    (Draft Bill, section 4(11), (12))
    To which court should the application be made?

    3.49         Should the application to have an acquittal set aside as tainted, and for authority to bring a new prosecution, be made to a single judge of the High Court or to a quorum of three judges of that court? PW Ferguson QC, both in his response to the Discussion Paper and in an article in Scottish Criminal Law,[96] suggests that any such application should be made to a single judge of the High Court, with each party having a right of appeal without leave.

    3.50         The existence of a right of appeal is certainly desirable, all other things being equal; but a number of considerations incline us in favour of requiring the application to be made to a quorum of the High Court. Principal among these is the desire to mark the serious and exceptional nature of the procedure. There is an obvious comparison with appeals against conviction, where a quorum is required to set aside a conviction. Although a decision to set aside an acquittal will not have any direct effect upon the liberty of the acquitted person, we do not think that the procedure for setting aside an acquittal should be any less onerous than that which applies in appeals against conviction.

    3.51         We recommend:

    20. Any application to have an acquittal set aside as tainted and authority granted for a new prosecution should be made to a quorum of three judges of the High Court.
    (Draft Bill, section 4(6))
    Fact-finding

    3.52         We are conscious that the procedures that we have proposed will commonly require some degree of fact-finding. This may not be required in cases in which some person has been convicted for interfering with jurors, but in cases involving interference with witnesses there will necessarily be a need for some form of factual inquiry into the evidence that would have been led but for the interference. Where the Crown does not rely upon a conviction to establish the commission of the offence against the administration of justice, there will have to be a hearing to establish the fact of the commission of the offence. We consider that the appropriate course in such cases would be for the application to be made to a quorum of the High Court, which could remit the application to a single judge who would make the necessary investigations into the facts. We consider that the detail of this process may safely be left to the High Court to prescribe by Act of Adjournal, but that it may be appropriate to make specific provision for the appointment by the single judge of an amicus curiae to act as a contradictor in cases where the accused person chooses not to oppose the Crown's application.

    Power to refuse application in the interests of justice

    3.53         In the Discussion Paper, we noted that one of the important safeguards built into the tainted acquittal procedure under the Criminal Procedure and Investigations Act 1996 was that the High Court was required to refuse to grant an application by the prosecution where to do so would be contrary to the interests of justice. We asked[97] whether a Scottish court considering a similar application should also have the power to refuse the application where such a refusal appeared to be in the interests of justice. Respondents agreed that it should.

    3.54         We recommend:

    21. The court considering an application under the tainted acquittal procedure should refuse that application where it appears to the court that to retry the accused would be contrary to the interests of justice.
    (Draft Bill, section 4(15))
    Reporting restrictions

    3.55         There is clearly a risk that a jury which learned that the accused had previously been acquitted following interference with witnesses or jurors would thereby be prejudiced against that accused. We asked whether it should be possible to mitigate this risk by allowing the court to impose restrictions on the reporting of the application. Respondents agreed that it should be possible to impose such restrictions.

    3.56         We recommend:

    22. The court hearing an application under the tainted acquittal procedure should have the power to impose restrictions upon the reporting of the application, with a view to preventing the publication of such reports from influencing the jury in any subsequent trial.
    (Draft Bill, section 11)
    Limitations on scope
    By type of proceedings

    3.57         In the Discussion Paper, we asked[98] whether the tainted acquittal procedure should apply to all offences, or whether it should be limited. We suggested that there was no reason in principle why there should be any restriction on the cases to which the procedure might apply, since it was important that the administration of justice should proceed properly in relation to less serious as well as more serious offences, but proposed that if some limitation were nevertheless thought appropriate, the procedure might be limited to cases originally tried in the High Court. Two respondents suggested that the procedure should be limited to cases originally tried on indictment, whether in the High Court or before a sheriff and jury. The remainder agreed with our principal suggestion that it was no more acceptable in principle for a person accused of a minor offence to avoid facing a fair trial than it was for a person accused of a more serious offence. The Judges of the High Court of Justiciary said:

    "We do not believe that this exception should be limited only to cases in the High Court. At whatever level within the criminal justice system an offence against the administration of justice occurs which has an effect upon the outcome of a trial it is a very serious matter. Thus we are of the view that at whatever level within the criminal justice system an offence occurs, the scope of the tainted acquittal exception should be wide enough to allow the consideration of a second trial.
    We do not believe that there would be such a body of cases brought within the said exception that there would be practical considerations justifying a limitation in the scope of the exception such as confining it to cases which originated in the High Court of Justiciary."

    3.58         We recommend:

    23. The application of the tainted acquittal procedure should not be restricted by reference to the court (solemn or summary) before which the original proceedings took place.
    (Draft Bill, section 4(1))
    Retrospectivity

    3.59         As noted above, it is our view that a retrial following a tainted acquittal does not constitute a real exception to the rule against double jeopardy. The underlying justification is that the original trial was not a proper trial, because of the interference with the process. Recommendation 13 above is little more than a statutory means of doing what appears to have been done in some of the cases to which Hume refers.[99] So, at first sight, there would appear to be no reason why the change should not apply to proceedings completed before any new legislation comes into force.

    3.60         We note, however, that the Criminal Procedure and Investigations Act 1996, which introduced a tainted acquittals procedure in England and Wales, applies only prospectively, in relation to cases where the offence of which the person was first acquitted is alleged to have been committed after the coming into force of the relevant provisions of the Act. The Law Commission, in discussing the question of retrospectivity in relation to exceptions to the double jeopardy rule,[100] noted that the tainted acquittals procedure in England and Wales had been made prospective only. They considered that that was a correct decision. At paragraph 4.54 they said:

    ". . . 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."

    3.61         We do not entirely follow the logic of that passage. The point of the tainted acquittal procedure is that someone whose trial has been corrupted by some criminal action should be retried on a proper basis. If it is possible to bring home to him, or to someone else, the commission of an offence of interfering with the administration of justice, then that is a separate offence. If he is convicted of an offence against the administration of justice, and punished, that is a separate punishment for a free-standing, separate offence. If that process is then followed, as in the English system, by an uncorrupted trial for the original offence, that uncorrupted trial does not represent an increase in the maximum sentence for the administration of justice offence, nor is any punishment attributable to that latter offence.

    3.62         In any event, what we are proposing in Scotland is that, before an application for a retrial is granted, it should be necessary only to prove that an administration of justice offence has been committed, without also securing the conviction of anyone for that offence. In such a case there could be no question of increasing the maximum penalty for an offence. Even where it did prove possible to convict someone of an administration of justice offence, we would not see that as making it incompetent to mount a second prosecution on the original charge. We would accordingly have no difficulty in principle with making the provisions relating to tainted acquittals retrospective.

    3.63         We recommend:

    24. The application of the tainted acquittal procedure should not be limited by reference to the date of commission of either the original offence, the offence against the course of justice which triggers the application of the procedure, or to the date of the allegedly tainted acquittal.
    (Draft Bill, section 4(2))
    Part 4 Should there be a "new evidence" exception?
  6. Introduction

    4.1         The previous Parts of this Report have considered whether retrial should be permitted where the original proceedings, although apparently valid, have in fact been tainted by some interference with the judge, jury or witnesses. In the context of the general rule against double jeopardy such a possibility would be a recognition that the original proceedings did not in fact constitute a fair trial. Accordingly, the accused was not in fact in jeopardy, and was not therefore entitled to the protection of the rule against double jeopardy. Such an innovation would not, therefore, be an exception to the rule. In this Part we wish to consider whether there should be any exception properly so called to the rule, first, where the accused has himself admitted committing the offence and, second, more generally.

    Retrial following admission by acquitted person

    4.2         Before considering whether there should be some general exception to the rule against double jeopardy, it would be sensible to look first at whether, in the particular case of an admission by the acquitted person, a more limited exception might be desirable. As we noted in the Discussion Paper,[101] such a confession seems to be qualitatively different from other forms of possible new evidence. If it is credible, it indicates that one of the parties to the original trial is admitting that it was conducted by him or on his behalf on a wrong basis. There is something profoundly disquieting about the notion that a person should effectively be able to boast - with impunity - that he has "got away with it".

    4.3         In some cases the law has addressed the matter. Most common law systems have allowed for the prosecution of an acquitted person for perjury, where he has denied committing the original offence in evidence at his trial, and has subsequently admitted having carried out the crime. That was also the case in HMA v Cairns, in which the court held that such a prosecution did not constitute a breach of the rule against double jeopardy. As Lord Wheatley put it:

    "The Lord Advocate summed up the situation by saying that where a supervening event takes place which could form the subject of a radically different charge, then that latter charge can be competently laid notwithstanding that this involves inquiry into the previous crime for which the panel has been tried and either convicted or acquitted. In my opinion the Lord Advocate's argument was well-founded and correctly represents the legal position."[102]

    4.4         The possibility of trying an acquitted person for perjury may have gone some way to assuage public concern at the prospect of such a person's being able, with impunity, to admit having actually committed the crime of which a jury had acquitted him. Similar decisions were reached, as we outlined in Appendix 2 to the Discussion Paper, in several common law jurisdictions. Indeed, we observed that it was the decision of the Australian High Court in the case of R v Carroll,[103] that such a trial would be an abuse of process since it would inevitably involve an attempt to controvert a murder acquittal, which prompted the reform of double jeopardy law in that jurisdiction. While the possibility of a trial for perjury - whether or not it is followed by a retrial for the original offence - may have gone some way to meet public concern at the prospect of an acquitted person's later admitting having committed the crime, that possibility is clearly not open where the acquitted person has not given evidence at the original trial. Nor is it easy to see why the possibility of a retrial should depend on there having been a previous conviction for perjury.

    4.5         Of those who responded to the Discussion Paper, only Professor Roberts was in principle against allowing a retrial where the acquitted person had subsequently admitted guilt. He pointed out various cases in which confession evidence had later proved to be unreliable, with the result that innocent people had been convicted, and drew attention, as did others, to the need for great care in accepting such evidence.[104] It is of course the case that courts are properly cautious in accepting such evidence, and it is noticeable that Miell[105] was a (particularly unconvincing) case of an alleged religious conversion. On the other hand, the two cases in which evidence of a subsequent admission has been accepted as new and compelling, Dunlop[106] and Celaire,[107] were instances of admissions to persons other than the authorities. They were also cases in which there was a considerable amount of other evidence to corroborate the evidence of the admissions.

    4.6         In our view an accused person who has credibly admitted that he is guilty of the offence of which he has been acquitted foregoes the equitable justification for the plea of res judicata which he could otherwise put forward as a bar to the subsequent proceedings. We are satisfied that, provided the courts continue to treat evidence of alleged admissions with care and that the usual requirements for corroboration remain in place, there is no reason why there should not be an exception to the rule against double jeopardy where a person has subsequently admitted committing the crime of which he has been acquitted. Such an exception would be justifiable and appropriate even if no more general exception were to be made.

    4.7         We recommend:

    25. It should be possible, with the authority of the High Court, to reprosecute a person who confesses to having committed an offence of which that person has previously been tried and acquitted.
    (Draft Bill, section 6)
    26. Application for authority to reprosecute should be made by the Lord Advocate to a quorum of three judges of the High Court.
    (Draft Bill, section 6(3), (6))
    27. The High Court should grant the application only if satisfied-
    (a) that the acquitted person has admitted having committed the crime to which the acquittal relates;
    (b) that the admission is credible and reliable; and
    (c) that it would not be contrary to the interests of justice to reprosecute the acquitted person.
    (Draft Bill, section 6(7), (8))
    Retrospectivity

    4.8         There are a number of general reasons to be cautious about retrospective legislation. The usual reason for caution with regard to such legislation is that it is seen as unfair and potentially oppressive to make laws which alter the basis upon which citizens have been conducting their affairs - certainly where the alteration is adverse to the citizens concerned. In the criminal sphere, Article 7 of the European Convention on Human Rights ("ECHR") prohibits not only the criminalisation of conduct which was not criminal at the time when it was committed, but also the imposition of a more severe penalty than was competent at that time.[108] More narrowly, one might argue that the introduction of any retrospective exception to the rule against double jeopardy would deprive an acquitted individual of a vested right or interest - namely the right not to be tried again for the same offence - and that this would be unfair. Finally, one might view the introduction of a retrospective exception as incompatible with the public interest in maintaining the finality of court judgments.

    4.9         None of these possible objections to retrospectivity applies with any force to the case in which a person who originally pled not guilty subsequently admits his guilt of the crime of which he was acquitted.  Allowing for a retrial in such circumstances does not criminalise conduct which was not criminal when committed.  Nor does it increase the maximum sentence applicable to that crime.  It merely makes it possible for the acquitted person who later confesses to be convicted and punished for precisely the crime with which he was originally charged.  There is no issue of incompatibility with Article 7 of the ECHR.

    4.10         Nor does either of the more general objections to retrospective application apply here.  Even if it is a valid objection to retrospective application of an exception that it would unfairly deprive the accused person of a settled right not to be tried again, we cannot see how this objection could apply to a person who has admitted his guilt.  It seems to us to be reasonable to regard such a person as having waived any such right.  He does not, of course, waive his right to be presumed innocent, and such cases may involve very difficult questions of credibility and reliability of evidence, particularly where the accused denies that he has made the admission, or that the admission, if made, was true. But where the fact and reliability of the admission can be established to the satisfaction of a court, the result is that the accused has at his own hand negated the basis upon which he approached the earlier prosecution, namely that he was innocent of the crime with which he was charged. We see no reason in principle why provision to deal with such a person should not be retrospective.

    4.11         We recommend:

    28. The application of the exception for post-acquittal confessions should not be limited by reference to the date of commission of the offence of which the accused was originally acquitted or by the date of the verdict of acquittal.
    (Draft Bill, section 6(2))
    A general exception for new evidence?

    4.12         We now turn to deal with the more general question: whether there should be an exception to the rule where there has been no interference with the course of justice, and no subsequent admission by the accused. The original prosecution was undertaken on the basis of evidence which the Crown considered to be sufficient, the trial process was carried out without interference, and the court has returned a verdict of acquittal. At some time after the trial, new evidence emerges which casts doubt upon the correctness of the original verdict. The question is whether that new evidence justifies re-opening the earlier proceedings. This is certainly the most critical, and the most difficult, issue thrown up by this reference. If introduced, such an exception would effectively remove the recognition by the law that a person who has "tholed his assize" cannot be prosecuted again for the same offence. However narrowly drawn it may be, any exception to the general rule against a retrial can be seen as ending the security of all those acquitted of crimes to which the exception relates.

    4.13         It was for that reason that we suggested that:

    "it would be necessary to find some substantial justification for such a departure from the present position: some compelling evidence that significant numbers of factually guilty persons are escaping justice because of the rule."[109]
    We indicated that we were unaware of any modern Scottish case in which the rule had operated so as to prevent the bringing to justice of a criminal, but invited comment on that question.[110]
    Responses to Discussion Paper

    4.14         No evidence or example of cases in which the rule has prevented the bringing to justice of a guilty person has emerged from the consultation process. The Association of Chief Police Officers in Scotland mentioned no such case. The Judges of the High Court of Justiciary were divided on the question of whether a new evidence exception should be introduced; but those who were against the introduction of an exception indicated that:

    "The members of this group are unaware of a body of cases in Scotland where the introduction of such an exception it is thought would give rise to retrials."
    COPFS, who were in principle in favour of an exception, were similarly unable to point to any existing cases in which a new evidence exception might operate, but suggested that the reason for this was that:
    "there has been little incentive to re-look at cases where there have been acquittals on the basis of the double jeopardy rule as it exists at present."

    4.15         None of the other respondents instanced a case where changing the rule would enable a prosecution to be undertaken in the light of new evidence. So far as Scotland is concerned, therefore, there is no evidence of any case where a person who may in fact be guilty is escaping justice because of the operation of the rule against double jeopardy.

    4.16         Nor does the evidence from England and Wales support the view that a substantial number of mistaken acquittals would be corrected by the introduction of a new evidence exception. During the passage of what became Part 10 of the Criminal Justice Act 2003, evidence was given to the House of Commons that there were some 35 cases which the police would be interested in re-opening if the law were changed.[111] In a very full and helpful response to our Discussion Paper, the Director of Public Prosecutions acknowledged that the figure of 35 cases mentioned in evidence to the House of Commons had not been borne out in practice. At the time of writing, more than four years after the coming into force of Part 10 of the 2003 Act, only 6 applications for a retrial have been determined. Three of the applications have failed.[112] Of the three successful applications only one did not relate to an admission by the acquitted person. Even in a much larger jurisdiction than Scotland, the absolute number of cases to which the exception has been applied is very small. The number is even smaller as a proportion of the cases to which the exception might in principle have applied: between 2001 and 2007 (the last year for which figures are available) there were more than 20,000 acquittals in relation to offences of a type to which Part 10 of the 2003 Act might apply. Even if Part 10 were not retrospective, the applications so far made would represent a tiny fraction of the number of acquittals which are in theory open to challenge under the new legislation.

    4.17         This can be seen in two ways. On the one hand, it may be an indication that the new provisions are being used carefully and discriminately by both the Crown and the Courts, and are therefore not a major departure from the rule against double jeopardy. Certainly, on a rational basis, none of those acquitted of one of the offences to which the legislation applies should feel that it is likely that the matter will be re-opened. On the other hand, it is in principle now possible that any of those acquittals may be re-opened. The finality and the security which was previously felt by all acquitted persons have been ended, in exchange for the conviction of a very small number of serious offenders.

    Arguments in favour of an exception

    4.18         Despite the lack of evidence of any cases in which a new evidence exception would enable the correction of erroneous acquittals, the great majority of respondents were in favour of introducing such an exception.

    4.19         A number of grounds were advanced in support of introducing a new evidence exception. These might be analysed under five headings: consistency; balance; scientific and technical advance; public perception; and moral principle.

    Consistency

    4.20         Those of the Judges of the High Court of Justiciary who were in favour of an exception indicated that:

    "[L]ogically if there should be an exception regarding later confession evidence there should be an exception for new evidence subject to proper safeguards as such admission evidence is merely a particular form of new evidence."

    4.21         PW Ferguson QC also considered that if an exception for post-acquittal confessions were to be introduced, then there should logically be a broader exception for new evidence:

    "It is for most people, morally speaking, unacceptable that persons who have been acquitted thereafter admit their guilt. Is it sufficient to say that these persons can be liable under the separate exception [for post-acquittal confessions]? Does not the answer to that question require the possibility of retrial where there is any new evidence of guilt of a compelling nature? It is surely irrelevant that in the latter case the accused may have bragged about 'beating the system'. If it is morally right to retry an accused because of new evidence coming to light after the original proceedings are concluded by acquittal, why should it matter where the new evidence comes from?"

    4.22         There is a significant distinction, however, between evidence of a post-acquittal confession and other new evidence. The justification for any exception to the rule against double jeopardy must involve, among other factors, a weighing of the positive value of the exception in allowing the conviction of the guilty against the general loss of finality. Under an exception for post-acquittal confessions, the acquitted person who subsequently admits, or confesses, or boasts, that he committed the crime of which he was acquitted thereby forfeits the equitable protection which he would otherwise enjoy against further prosecution. The acquitted person who does not later confess retains the protection of the rule against double jeopardy. The exception to the finality of criminal judgments is strictly limited. When viewed from this perspective, there does not appear to be any logical inconsistency between allowing an exception based upon a post-acquittal confession and refusing to allow such an exception on the basis of other new evidence.

    Balance

    4.23         Another respondent[113] suggested that since the options for challenging a conviction had been expanded, particularly by the introduction of the Scottish Criminal Cases Review Commission, it would be appropriate to introduce a new evidence exception in order to rebalance the system towards the interests of victims and those of society in convicting the guilty:

    "A modern criminal justice system requires to balance the interests of the accused with those of any victims and the general public. Since 1999 any person convicted of an offence in Scotland, whether solemn or summary, is entitled to have a conviction reviewed at any time if he or she believes that there may have been a miscarriage of justice . . . The consequence of the creation of the SCCRC . . . is that the notion of there being finality in criminal procedure is now something of a myth, as any witness, either for the Crown or the defence, can be contacted by the SCCRC . . . no matter how long after the trial. In order to achieve an appropriate balance there should be certain limited circumstances where an exception to the rule against double jeopardy should be allowed, and new evidence which could be considered both material and compelling would be one such exception."

    4.24         We do not find this argument persuasive. Of course the interests of victims and society in bringing the guilty to justice are important. If we did not value such interests highly, we would not even be considering introducing exceptions to the rule against double jeopardy. But as we have observed elsewhere in this Report, the whole basis of our criminal procedure is asymmetrical, preferring to accept the acquittal of some guilty people rather than risk condemning more of the innocent. Viewed with this in mind, the introduction of new means of identifying cases in which justice has miscarried by convicting the innocent does not imply that we must introduce counterbalancing changes to favour the prosecution.

    Scientific and technical advance

    4.25         More than one respondent referred to scientific advances such as DNA evidence as justifying a new evidence exception. The Faculty of Advocates was of the view that:

    "[i]n light of advances in forensic science, medical science and technology, and the possibility of subsequent confession discussed above, there should be an exception to the rule against double jeopardy on the basis of new evidence."

    4.26         Although a number of respondents made the point that new evidence might become available as a result of technological advances, and the prospect of such advances formed part of the justification for introducing the exception in Part 10 of the 2003 Act, we note that none of the cases so far decided under the 2003 Act has depended upon new technology.

    4.27         New scientific and technical ways of analysing physical evidence may be relevant in one of two ways. Either existing developments in forensic science - principally DNA analysis - might be applied to cases which were originally heard before such technology was developed, or new technology might be developed which would enable new evidence to be produced in cases which were first tried after the exception is introduced.

    4.28         The value of forensic science techniques such as the analysis of blood groups, fingerprints and the various forms of DNA analysis lies in their ability to tie samples of physical evidence taken at the scene of a crime with samples taken from suspects. An obvious point, sometimes overlooked, is that in order to apply such techniques there must first be suitably preserved samples from the crime scene. In unsolved cases, such samples are routinely retained and there have been a number of cases in which historic cases have been prosecuted on the basis of DNA evidence. Such cases are not without their difficulties - in particular, there may be issues regarding contamination of samples which were collected before DNA analysis was common and which may have been stored in inadequate conditions - but here DNA evidence is useful.

    4.29         It is doubtful, however, whether such evidence is likely to be of similar use where there has already been a trial. We understand from our discussions with COPFS that, following an acquittal, productions are routinely destroyed. It seems unlikely, therefore, that the existing techniques of DNA analysis or technological advances, in the sense of technology which is developed and becomes available after the conclusion of the proceedings, will be of much assistance in relation to cases concluded before any putative rule change.

    4.30         We might also observe that it is inherently unlikely that the existing techniques of DNA analysis, impressive as they are, would generate new evidence in cases which were tried after the introduction of a new evidence exception. DNA analysis might assist in demonstrating the significance of newly-discovered physical evidence - perhaps in a murder case in which the body was not found until after the original murder trial - but could not in itself generate new evidence, since DNA techniques would have been available at the time of the original prosecution.

    4.31         The routine disposal of productions in cases which result in acquittal means that neither DNA nor any new forensic technique is likely to produce new evidence in historic cases, and the present availability of DNA evidence means that DNA is unlikely to produce "new" evidence in future cases.

    4.32         If a new evidence exception were to be introduced, it would obviously be possible for productions to be kept routinely. We have not investigated the physical and financial implications of such a practice, although, having regard to the English experience,[114] it would appear that an immense number of productions would require to be kept for very little practical result. Nevertheless, it might well be that, if the productions in every case of serious crime were routinely preserved, in conditions such as to prevent contamination from contact with other materials, it would be possible, when some new technology was developed, to test those productions in the light of the new knowledge. The practical benefit of an exception may depend, in large part, upon the likelihood of new forensic science techniques being developed which could wring new evidence from existing samples. It has been suggested that future developments might include "scientific advances . . . on finger-printing, photographic recognition, corneal and face mapping and even reliable ear-prints."[115] The likelihood of such techniques yielding significant and reliable evidence is a matter of speculation,[116] although even the most optimistic supporter of new forensic science must admit that their potential value does not approach that of fingerprinting, blood-typing or DNA.

    Public perception

    4.33         Other respondents regarded a new evidence exception as justified by a need to maintain public confidence in the system of criminal justice. Dr Campbell and Professor Duff, of Aberdeen University, suggested that:

    "[T]he criminal justice system must retain the support and respect of the public. For this reason, it is desirable that if compelling new evidence emerges of an acquitted person's guilt, a fresh prosecution is possible."
    They went on to suggest that if such an exception were to be introduced, it were well that it were done in a considered manner and not as a knee-jerk reaction to some high profile case:
    "Any such legislation must be tightly framed. In this context, to take action now avoids a possible future scenario where there is public outrage over a notorious case in which double jeopardy prevents such a prosecution and the result is the rushing through Parliament of hastily prepared and unsatisfactory 'knee-jerk' legislation."

    4.34         Although we are unaware of any present case in which the double jeopardy rule is preventing a conspicuously guilty person from being brought to justice, it is not difficult to imagine such a high profile case arising in the future. One obvious scenario would be a murder prosecution in which the body of the victim was not discovered, despite thorough inquiries, until after the trial and acquittal of the alleged murderer. If the body, when discovered, yielded clear evidence of the guilt of the acquitted person, one might readily imagine a public outcry if the rule against double jeopardy prevented that person from being tried again.

    Moral principle

    4.35         Of course, if one were of the view that it would always be unjustified, whether morally or politically, to allow for an exception to the rule against double jeopardy on the basis of new evidence, then this would be an end of the question, regardless of the possibility of outraged public opinion. But there is an undeniable value in bringing those guilty of serious crimes to justice, and it may not be unreasonable to regard this as being capable of outweighing the general interest in the finality of acquittals in some limited circumstances. The moral justification for a new evidence exception was advanced by PW Ferguson QC, who said:

    "The justification for making such an inroad on the protection [against double jeopardy] must ultimately be a moral one: that justice has miscarried in the original verdict and that it is therefore just that the accused be retried. If that moral justification is accepted, I wonder whether it matters that there is only evidence of one case rather than 10 or 100 where retrial would be in order. Indeed, for there to be a change must there be evidence of even one case meriting retrial if change is morally warranted?"

    4.36         In the absence of any evidence of outstanding cases in which the introduction of a new evidence would allow the prosecution of arguably guilty people who had escaped justice because of the rule against double jeopardy, the general question, as to whether it would be sensible to have an exception to the rule against double jeopardy, must be approached on the basis of the competing public interest issues. It is therefore appropriate briefly to re-state the arguments underlying the basic rule against double jeopardy.

    Reasons for the rule against double jeopardy

    4.37         We have identified three principal reasons for the rule against double jeopardy. The first is that identified by Hume, that, after an acquittal:

    "the pannel can never again be challenged or called in question, or made to thole an assize . . . on the matter or charge which has been tried. The ground of which maxim lies in this obvious and humane consideration, that a person is substantially punished, in being twice reduced to so anxious and humiliating a condition, and standing twice in jeopardy of his life, fame, or person."[117]

    4.38         The second reason is the general public interest in securing finality in legal proceedings, so that the parties to the dispute, and others less immediately affected, can carry on with their lives and business on the basis that the court's decision has settled the issues between them, and so that the (scarce) resources of the judicial system are not expended in the rehearing of the same issues between the same parties. If it were not so, then a litigant with a particular grievance against another citizen, and the funds to carry on litigation, could repeatedly raise the same issue before the courts. That principle is perhaps even more important in relation to criminal matters than it is in civil affairs, because in criminal matters the pursuer is the State, with its comparatively vast resources. As Black J expressed the matter in the United States Supreme Court:

    "The underlying idea, one that is deeply ingrained in at least the Anglo-American system of jurisprudence, is that the State with all its resources and power should not be allowed to make repeated attempts to convict an individual for an alleged offense, thereby subjecting him to embarrassment, expense and ordeal and compelling him to live in a continuing state of anxiety and insecurity as well as enhancing the possibility that even though innocent he may be found guilty."[118]

    4.39         That formulation reflects not only Hume's idea of protecting the individual from repeated personal exposure to the risks of the criminal process, but also the more general public interest in the finality of court decisions. It may also be seen as reflecting what we would see as the third rationale for the rule, which is that it is an affirmation of the fundamental constitutional status of the individual as against the State. The individual is presumed by the law to be innocent until proven to be guilty. If the State has grounds for challenging that presumption in relation to a particular alleged offence, it may charge him with it in accordance with the prevailing rules. If the individual is acquitted after a proper trial, then the State should not be permitted to repeat the process.

    4.40         Although a clear majority of respondents were in favour of introducing a new evidence exception, not all respondents were in favour, and dissenting voices ranged from doubt to unequivocal opposition. Mr MGA Christie, formerly of Aberdeen University, said:

    "This is, I think, a difficult matter. On the one hand, the public interest would seem to be served by the hearing in court of new and compelling evidence; on the other, no acquitted person would ever feel secure if he might be retried at any future time on the basis of such evidence. On balance, I am inclined to favour a time limit . . . beyond which no exception based on new evidence would apply . . . or an outright rejection of the exception."

    4.41         In the same vein, those of the Judges of the High Court of Justiciary who were against an exception said:

    "An exception of this type would in a real sense render all acquittals as not being final determinations, which is one of the main reasons for the rule against double jeopardy."

    4.42         Professor Paul Roberts regarded a new evidence exception as wrong in principle. He strongly endorsed the view that any such exception could only be justified by persuasive evidence that it would be likely to enhance the cause of justice, and suggested that if no such evidence were produced in response to our Discussion Paper (and none has been) then the argument for a new evidence exception could not possibly be made out. As well as founding upon the lack of evidence of practical benefit from the introduction of an exception, Professor Roberts expressed fundamental opposition to an exception as a matter of principle:

    ". . . I would want to say that the double jeopardy prohibition absolutely, without exception, prevents repeat prosecutions for the same wrongs. If the justification for having the prohibition lies in the support it lends to personal autonomy, political security and democratic accountability, then these values are maximally promoted by an absolute prohibition that says: once tried to a final conclusion, the matter is at an end." (emphasis in original)
    Conclusion

    4.43         We have considered these issues carefully, both from first principles and in the light of the responses to our Discussion Paper. There are strong arguments on both sides, and a division of opinion within the Commission. We therefore make no recommendation as to whether or not there should be an exception to the rule against double jeopardy on the basis of new evidence.

    4.44         But, as we pointed out in the Discussion Paper, and as is indeed clear from the responses we have had, there is a considerable body of opinion which considers that there should be a carefully drawn and narrowly defined exception to the rule against double jeopardy. Part 5 of this report considers how any such exception might best be formulated.


    Part 5 The formulation of a "new evidence" exception
  7. Introduction

    5.1         This Part considers the technical aspects of a new evidence exception, should it be decided that such an exception be introduced. It does not directly address the merits and demerits of the introduction of such an exception; these are questions which are addressed in depth in Part 4 above.

    To what offences might a new evidence exception apply?

    5.2         In the Discussion Paper, we noted that when the issue of a new evidence exception was considered in England and Wales, the Law Commission recommended that the exception should apply only to a very limited range of offences.[119] The Criminal Justice Act 2003, which introduced a new evidence exception in England and Wales, applied the exception to a considerably wider range of offences, specified by name in a Schedule to the Act.[120] We noted also that the common law basis of the majority of serious offences in Scotland, together with the lack of a set sentencing tariff, meant that it was difficult to define a category of offences to which an exception should apply by reference to the seriousness of the offence, or to the punishment which could be imposed. We suggested, however, that it might be appropriate to confine any exception based upon new evidence to the most serious crimes, and asked whether it would be appropriate for such an exception to be limited to offences which were originally tried in the High Court of Justiciary.[121]

    5.3         This question provoked a range of responses. A number of respondents agreed that it would be appropriate to limit the exception to cases originally tried in the High Court. Others pointed out that a number of very serious offences were tried before a sheriff and jury and suggested that a more appropriate limit might be to include only those cases which were first tried on indictment, whether in the High Court or before a sheriff and jury. Sometimes the effect of the new evidence might be to show that an offence which was originally tried on indictment before a sheriff and jury should now be tried before the High Court; COPFS gave the example of a case in which a new complainer emerges following the accused's original acquittal, where it should be possible to apply the Moorov doctrine. One respondent suggested that there should be no limit upon which offences might be retried, the matter being one which could safely be left to the discretion of the Lord Advocate. There was also some support for a list of specified serious offences.

    5.4         We consider that if a new evidence exception is to be introduced, it should apply only to the most serious offences. That is, any exception should be just that, an exception to be applied in strictly limited cases, and not merely the abolition of the rule against double jeopardy. The limitation of the exception is important, since it minimises the extent to which those once tried for an offence are at risk of subsequent prosecution. The narrower the range of offences to which the exception might apply, the less general the harm to the finality of verdicts in general.

    5.5         For this reason, we reject the suggestion that a new evidence exception should apply in relation to any offence which was originally tried upon indictment. The sentencing power of the sheriff in solemn cases is limited to five years' imprisonment. We would not seek to deny the seriousness of offences tried before a sheriff and jury. It is clear, however, that the punishment that can be imposed under the limited sentencing powers of the sheriff is far below those levels of punishment imposed for the most serious crimes. If a new evidence exception should apply only in the most serious cases, it is reasonable to exclude any case which the prosecutor has deliberately chosen to bring before a court with such limited sentencing powers.

    5.6         How could the exception be restricted to the most serious cases? There is no direct, objective, and uncontroversial test of seriousness. On one view, the only principled point at which to draw a line is between murder - the only offence to attract a mandatory life sentence[122] - and other offences. That would be a legitimate restriction of what is to be a limited exception. Against that, we are conscious that life imprisonment for murder has been refined into the punishment part of the sentence which is, in very many cases, then followed by release on licence. In its effect, that regime has similarities with the power of the court to impose extended sentences in relation to serious sexual offences (and offences of violence). We would accordingly add to the crime of murder the crime of rape, which is the most serious sexual offence. Since we appreciate that there is no logical or principled basis for this limitation, and in order to provide a means of adjusting the list in the future, we would suggest that Scottish Ministers should, subject to the approval of the Parliament, retain the right to add further offences to the list by order. Any such amendment should be of purely prospective effect, applying only to cases in which the verdict of the original trial is delivered after the amendment has come into force.[123] We propose:

    29. A new evidence exception should apply only in relation to the offences of murder and rape.
    (Draft Bill, section 8(9))
    30. Scottish Ministers should be able to add further offences to the list by way of affirmative order.
    (Draft Bill, section 8(10))
    The test of "newness"

    5.7         By definition, a new evidence exception must rely upon new evidence. It could never be acceptable to allow someone to be tried repeatedly on the basis of the same evidence: this would be utterly incompatible with the need for finality in criminal justice. But what, exactly, counts as "new"?

    New evidence in England and Wales

    5.8         Before considering the test which should apply in Scotland, it is instructive to consider the provisions of the new evidence exception in England and Wales, as set out in section 78 of the Criminal Justice Act 2003.

    5.9         Two things should be noted about the definition of "new" evidence in section 78. The first is that there is no requirement that the evidence should not have been available, or available with the exercise of reasonable diligence, at the original trial. In order to satisfy the definition of newness, it is sufficient that the evidence was not adduced in the original proceedings;[124] whether or not it would have been so adduced had it not been for the failure of the prosecution to act with due diligence is merely a factor to be taken into account by the Court in determining whether the making of an order would be in the interests of justice.[125]

    5.10         In response to our Discussion Paper, the Director of Public Prosecutions, Keir Starmer QC, confirmed that it is the policy of the Crown Prosecution Service to seek to quash an acquittal only on the basis of new evidence which was not in the possession of the prosecution at the original trial, or, if it was in their possession, the probative value of which could not objectively have been appreciated at the time because of, for example, limitations in forensic science. Nevertheless, the terms of the 2003 Act would permit the prosecution to seek to have an acquittal overturned on the basis of evidence which was in the possession of the prosecution, but was simply not led for tactical reasons. Nor does the 2003 Act contain a rule preventing reliance upon evidence which the prosecution should, with reasonable diligence, have had available at the original trial, preferring to regard this as merely one of a number of factors to be taken into account in applying the section 79 interests of justice test.

    5.11         As we noted in the Discussion Paper,[126] the European Court of Human Rights, in construing Article 4 of Protocol 7 to the European Convention on Human Rights, has held that errors by the prosecution authorities or the courts cannot be used to justify the reopening of a case against an individual. As the Court remarked in Radchikov v Russia:

    "The Court considers that the mistakes or errors of the state authorities should serve to the benefit of the defendant. In other words, the risk of any mistake made by the prosecuting authority, or indeed a court, must be borne by the state and the errors must not be remedied at the expense of the individual concerned."[127]

    5.12         Although the UK has not yet signed and ratified Article 4 of Protocol 7, the principle articulated by the court in Radchikov is a sound one. We consider that any legislation introducing a new evidence exception should go further than the 2003 Act in restricting the new evidence to evidence which the prosecution could not reasonably have led at the original trial.

    5.13         The second notable feature of the 2003 Act's definition of "new" evidence is that, in focusing upon the question of whether or not evidence was adduced in the original proceedings, it includes evidence which was in the possession of the prosecution at the time of the original trial but which was not then admissible. Changes in the law which alter the rules on the admissibility of evidence - perhaps by permitting evidence obtained from covert interception of communications, or allowing evidence of bad character to be used to show propensity - may enable the reprosecution of an acquitted person.[128]

    The need for reasonable diligence

    5.14         In the discussion paper we asked whether, for the purposes of a new evidence exception, "new" evidence should be evidence which was not, and could not with the exercise of ordinary diligence have been, available at the original trial. There was general agreement with this proposed test, although one respondent[129] suggested that the test to be satisfied by the prosecution should be the same as that which is faced by the defence in seeking to bring an appeal on the basis of new evidence in terms of section 106(3) of the Criminal Procedure (Scotland) Act 1995. As originally enacted, section 106 required that in order to form the basis of an appeal against conviction, new evidence needed to be evidence "which was not available and could not reasonably have been made available at the trial". Were this still the test, we would have had no hesitation in recommending that the same test should apply to both prosecution and defence. However, the present test under section 106 requires merely that the appellant provide a reasonable explanation of why the evidence was not heard at the original proceedings.[130] This provision is not as liberal as it appears, and has been interpreted by the Appeal Court as preventing evidence from being regarded as "new" which might, with due diligence on the part of the defence, have been led at the original trial.[131] It may thus be the case that the application of the existing section 106 test would place an appropriate limit on a new evidence exception. We think, however, that it would be appropriate to hold the prosecution to a higher standard, and that it would be appropriate for the test of "newness" to be set out in statute in plain terms.

    5.15         We recommend:

    31. For the purpose of a new evidence exception, evidence should be regarded as "new" only if it was not, and could not with the exercise of reasonable diligence have been, available at the original trial.
    (Draft Bill, section 8(7)(b))
    The effect of changing rules of admissibility

    5.16         Should it be possible to rely on evidence as "new" where that evidence was available at the time of the original trial, but could not then be led owing to a rule excluding such evidence? As we have already noted, the Criminal Justice Act 2003 allows evidence to be treated as "new" not only where it is newly available to the prosecution, but also where previously available evidence is newly admissible. The same is true of appeals on the basis of new evidence under section 106 of the Criminal Procedure (Scotland) Act 1995: inadmissibility of evidence at the time of the original trial may be a reasonable explanation for the defence's failure to lead such evidence, and the newly admissible evidence may be relied upon to found an appeal "if it appears to the court that it would be in the interests of justice to do so."[132]

    5.17         This is a difficult question. On the one hand, the inadmissibility of a certain type of evidence supplies the clearest possible reason for the prosecution's failure to lead that evidence at the original trial, removing any suggestion that the prosecution acted without due diligence, or in an improper attempt to manipulate the trial process, by withholding that evidence. On the other, allowing the prosecution to take advantages of changes in the law of evidence which expand the range of admissible evidence smacks of changing the rules after the game has already been played.

    5.18         The Law Commission also found this question to be difficult. They were initially in favour of allowing newly admissible evidence to be treated as new,[133] suggesting that the situation was analogous to one in which the prosecution authorities were aware of the existence of strong evidence at the time of the original trial, but were unable to find it. Respondents to the Law Commission's consultation paper roundly rejected this approach, arguing that there was a risk that the law of evidence would be changed in order to allow the reprosecution of particular cases.

    5.19         It would clearly be improper to change the law of evidence after a person had already been tried in order to allow the reprosecution of that person. If such changes in the law of evidence were to allow the reopening of past cases on the basis of the "new" evidence that they generated, there could be significant temptation to modify the rules of evidence in relation to particular high-profile cases. We think that this temptation is best avoided by limiting the scope of any new evidence exception to evidence which was not available to the prosecution at the time of the original trial.

    5.20         We recommend:

    32. Evidence which was available to the prosecution (or could, with ordinary diligence, have been available) at the first trial but which was then inadmissible should not be regarded as "new" by virtue of any subsequent change in the rules relating to the admissibility of evidence.
    The significance of the new evidence

    5.21         What standard should the court apply in deciding whether new evidence is sufficiently significant to warrant the setting aside of an existing acquittal and the granting of authority to retry the accused? In the Discussion Paper, we sought the views of consultees on the question of the appropriate standard, and stated our own provisional view that any application for a retrial "should be subject to consideration by the appeal court of the likelihood that the new evidence would substantially improve the chances of a conviction."[134]

    The overall strength of the case

    5.22         Responding to our suggestion, COPFS stressed the importance of considering the new evidence in the context of the case as a whole:

    "The evidence will require to be significant in the context of the case. If the evidence is credible and reliable and a reasonable jury would have been entitled to convict the accused on that evidence along with the other evidence in the case then such new evidence would justify the new trial."

    5.23         We do not think that this will do. Of course, this is a necessary condition for allowing a new trial - if the evidence, including the new evidence, would not be enough to allow a reasonable jury to convict, then there can be no question of it being appropriate to bring the case to trial. Yet it is not sufficient: on this test, providing that the original case against the accused was sufficient to go to the jury at all, any new evidence would justify a new trial. The test would pose a negligible hurdle in any case where the original case against the accused was strong, allowing new trials in any case in which the Crown was surprised by a not guilty or not proven verdict, subject only to the need to produce some minimal amount of evidence which satisfied the test of newness.

    Substantially improving the chances of conviction

    5.24         There was significant support among respondents for our suggestion that the new evidence must substantially improve the chances of conviction. We stand by that view: if the new evidence does not substantially improve the chances of conviction, it is hard to see how it could possibly justify a new trial. Nevertheless, we have come to recognise that this test does not capture all that it should. It fails in the opposite direction to the test proposed by COPFS: while that test would allow a retrial on the basis of very little new evidence if the original case was strong, this test would be easiest to satisfy where the original case was very weak. If the original case was very weak, almost any additional evidence might substantially increase its strength.

    5.25         A better test might incorporate elements of each of the preceding tests, requiring both that the new evidence substantially increase the chances of conviction and that the evidence, taken together with the evidence in the original case, would entitle a reasonable jury to convict.

    5.26         A similar hybrid approach was considered by the Law Commission in their Consultation Paper on Double Jeopardy.[135] The Law Commission proposed a two stage test. First, the prosecution would have to show that the new evidence made the case, as a whole, substantially stronger than it was at the first trial. Second, it would have to show that the likelihood of a conviction at a retrial is of some minimal level. They invited comment on what that level should be, suggesting two possible alternatives: either (a) that it is highly probable that a jury would convict; or (b) that the court is sure that a jury would convict. The Law Commission ultimately rejected both elements of its proposed test.

    5.27         With regard to the first limb - whether the new evidence produced a substantially stronger case - the Law Commission was impressed by the observation that the extent to which the case was strengthened could not properly be assessed until the evidence had been tested in court, and that the test could be difficult to apply and would be likely to lead to uncertainty and dispute. They also noted that while the test of substantial improvement prevented retrials from being authorised where the new evidence added only marginally to the cogency of the evidence adduced at the first trial, something which would amount to little more than a pretext for challenging the original jury's verdict, it had the corollary that we noted above: the test was more easily satisfied where the original case was weak. This the Law Commission regarded as a fatal flaw.

    5.28         The Law Commission also came to regard the second element of their proposed test as inappropriate. They noted that nearly all of their respondents had commented upon 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 speculate upon the outcome of a trial which had yet to take place, and that the knowledge that the evidence had passed such a test might prejudice any subsequent jury. The Law Commission thus rejected this approach, and sought a test which would not require the court to speculate about the likely outcome of a retrial.

    "Compelling" evidence

    5.29         The solution which the Law Commission proposed, and which ultimately formed the basis of sections 77 and 78 of the 2003 Act, was to consider whether the new evidence was "compelling". In the Law Commission's view, a retrial should be allowed only where the new evidence "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."[136] The Commission was at pains to point out that putting the new evidence in context did not mean considering that evidence in relation to the other evidence at the original trial, but rather considering it in the context of the issues that arose at that trial:

    "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 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 in 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."[137]

    5.30         The Law Commission's recommendation in this regard was implemented in the Criminal Justice Act 2003. Section 78(3) of that Act says that evidence is compelling if it is reliable, substantial and, in the context of the outstanding issues, it appears highly probative of the case against the acquitted person. The "outstanding issues" are the issues in dispute in the proceedings in which the person was acquitted.

    5.31         As we have seen, the Law Commission's stated intention in framing the test of compelling evidence was to avoid requiring the court to speculate on the outcome of a future trial, potentially usurping the role of the jury and prejudicing any future trial. Did they succeed?

    "Compelling" evidence in practice

    5.32         The Director of Public Prosecutions informed us of the policy of the Crown Prosecution Service regarding the interpretation of section 78(3):

    "The court will obviously only be interested in authorising re-trials in cases where they think the new evidence will have a substantial impact on the outcome. Given the fundamental importance of the rule against double jeopardy, the standard by which the appellate court judges that impact should, in my view, be very high indeed. For its part, CPS has consistently interpreted the requirements of section 78(3), which define when new evidence is "compelling", in a narrow way. It is our publicly stated policy that the DPP's permission to proceed with an application to quash an acquittal will only be given in cases in which, as a result of the new evidence, a conviction is highly probable, either by a plea of guilty or by the verdict of the jury, and any acquittal by a jury would appear to be perverse. The probative value of the new evidence in each case is to be assessed on its own merits. In addition, as the language of the Act makes clear, the new evidence must, of itself, be compelling: in other words, it is not sufficient merely to add that new evidence to the original case, to assess whether the case as a whole is now compelling."

    5.33         At the time of publication of the Discussion Paper, we were aware of only two applications under Part 10 of the 2003 Act: R v Dunlop[138] and R v Miell.[139] Each was a case in which a person acquitted of murder had subsequently confessed. As such, the new evidence certainly qualified, if reliable, as compelling in itself - if the confessions were believed, they would, on their own, have been sufficient to establish guilt.

    5.34         Pace the DPP, however, it does not appear that the new evidence need be compelling in itself, or merely in relation to the outstanding issues at the original trial, without further reference to the evidence which was actually adduced. In order to see how the "compelling evidence" approach has worked in practice, it is necessary briefly to consider the small number of cases in which the Court of Appeal has been called upon to authorise a new prosecution under Part 10 of the 2003 Act.

    5.35         In the first such case, R v Dunlop,[140] the application was successful. Some years after his original acquittal, Dunlop had confessed to murder and been convicted of perjury in relation to his original denial on oath. The case is of little interest for present purposes, since Dunlop did not withdraw his confession and no challenge was mounted to the prosecution's assertion that the evidence of his confession, and of his conviction for perjury, was compelling evidence of his guilt. We discuss the case of Dunlop further below in relation to the questions of whether any new evidence exception should have retrospective application, and in relation to the potentially prejudicial effect of a court's finding that new evidence is compelling upon the fairness of a subsequent trial.

    5.36         In R v Miell,[141] the second application made under Part 10 of the 2003 Act, the Court of Appeal refused to authorise a new trial on the basis of Miell's post-acquittal confession to murder and his subsequent conviction of perjury, on the basis that these confessions were not reliable. The court observed that "[i]f there were no discrepancies between the prosecution's evidence and Miell's confessions the latter would be overwhelmingly compelling that Miell was the man who murdered Mr Burton."[142] In deciding that Miell's confessions were not reliable, the court referred to the incompatibility of the detail given in those confessions with the forensic evidence relating to the murder weapon and the injuries of the deceased, and with the parole evidence of other witnesses.[143] In terms of section 78 of the 2003 Act, evidence must be reliable before it can be compelling, and the reliability of the evidence cannot be assessed except in the light of the other evidence in the case, not merely the outstanding issues. A confession, if reliable, will always meet the tests of being substantial and highly probative; but the issue of its reliability cannot be resolved in an evidential vacuum.

    5.37         The third application to be considered under Part 10 reinforces the view that the courts will in practice consider the evidence as a whole, rather than requiring the new evidence to be compelling in itself. In R v Andrews,[144] the respondent had been acquitted of indecently assaulting and then raping SN, a girl of 15 who assisted at a summer camp run by his company. Thirteen years had passed between the alleged offence and the trial. The evidence of SN, who also gave evidence of having suffered sexual abuse in other unrelated incidents, was largely unsupported. Andrews presented himself as a man of good character who had a long and unblemished record of working with children. The jury acquitted on all charges. Following the acquittal, Andrews' ex-wife read of the case and went to the police, informing them that he had been arrested, many years previously, in connection with the indecent assault of three children at the school at which he then worked. Following an extensive police investigation, a further indictment was laid against Andrews, charging 17 counts of offences relating to sexual offences against a number of youths in relation to whom he had enjoyed a position of responsibility and trust. The Crown Prosecution Service sought to have Andrews' acquittals of indecent assault and rape quashed, and a retrial granted on the charge of rape, on the basis of this new evidence.

    5.38         By virtue of the similar fact evidence provisions of the 2003 Act, the evidence of the other offences would be admissible to prove the charge of rape relating to SN. It was clearly relevant both to his credibility - a significant outstanding issue in the original case - and to his propensity to commit offences of a similar character. Despite this, we very much doubt if this evidence could be regarded as compelling in itself, since none of it related directly to the alleged facts of the offence committed against SN. In the absence of some such evidence, no amount of evidence showing that Andrews had a propensity to commit similar offences could be compelling evidence of his guilt of the particular allegation. Andrews' counsel argued, unsuccessfully, that since the new evidence did not relate directly to the facts of the alleged offence against SN, it was not evidence "in relation to the qualifying offence", and therefore did not fall within section 78(1). The Court of Appeal refused to accept that there was a distinction to be drawn between "direct" evidence which might trigger the new evidence exception and "indirect" evidence which would be incapable of doing so. The Court of Appeal held that the relevant question was, rather, whether the evidence was admissible to prove that, contrary to his evidence at trial, Andrews had raped SN:

    "What matters is that the evidence should be admissible to prove that, in accordance with her complaint, and contrary to his evidence at trial, the respondent raped her. It would be contrary to the purpose of the legislation for new, compelling, highly probative, admissible evidence that he did so to be disregarded. [Otherwise] we should end up with a new concept, that is two compartments, both containing evidence admissible in law to prove guilt if deployed at a second trial, but with evidence from one compartment excluded from consideration when addressing the question whether the acquittal should be quashed and a second trial ordered. In the context of the legislative purpose such compartmentalisation would be remarkable."[145]

    5.39         Remarkable or not, it does seem that a distinction such as that argued for by Andrews' counsel is essential to the Law Commission's proposed requirement that the new evidence should be compelling in itself, since if new evidence does not relate directly to the alleged facts of the offence it is hard to see how it could ever, in itself, be compelling evidence of the accused's guilt of that offence.

    5.40         The Court of Appeal also quoted with approval the DPP's statement that "he would only proceed in cases where, as a result of new evidence, a conviction is highly probable and any acquittal by a jury at a subsequent trial would be perverse," characterising this guidance as "entirely appropriate, and consistent with the relevant legislative framework." In the course of allowing the application, the Court said:

    "[The CPS] rightly contend that the new evidence shows that SN's allegation was not an isolated complaint against a man of good character who spent his adult life blamelessly working with children, but as now appears, one in a series of independent allegations forming a pattern of abuse of those in his care or for whom he was in a position of authority and trust. Even if not "direct" this provides strong supporting evidence for SN which was not available at trial, and the evidence that the respondent was guilty of the rape of SN is now significantly more powerful than it was. In our judgment, if it had been available at the first trial, or if it were now to be deployed at a second trial, the high probability is that the respondent would have been or will be convicted."[146]

    5.41         It is apparent from this passage from the Court of Appeal's judgment that the court does not consider whether the new evidence is compelling in isolation, or merely in relation to the outstanding issues at the original trial, but also takes into consideration both of the factors which were considered and rejected by the Law Commission, namely the extent to which the new evidence strengthens the original case and the likely outcome of a new trial at which the new evidence was led.

    5.42         Andrews was followed by the Court of Appeal in R v B(J).[147] There the question was whether the acquittal of B(J) could be set aside, and a retrial ordered, on the basis of evidence of one of the co-accused at his original trial who had, after being convicted at that trial, entered into an agreement under section 74 of the Serious Organised Crime and Police Act 2005[148] by which he hoped to secure a reduction in sentence by providing evidence incriminating B(J). The Court, after referring to the acceptance of the DPP's statement in R v Andrews, went on to stress that the test under section 78 required the Court of Appeal to consider not merely whether the new evidence was capable of being reliable, but whether it was reliable.[149] As in the case of Miell, the assessment of the reliability of the new evidence could not be considered except in the light of the evidence given at the original trial, in this case mediated by the conviction of the proposed new witness for involvement in the offence of which B(J) was originally acquitted. R v B(J) was followed in R v G(G),[150] which also involved an attempt to use the evidence of a former co-accused. The most recent successful application under Part 10 was the well-publicised case of R v Celaire,[151] in which the credibility of Celaire's post-acquittal confession can only have been enhanced by its having been given to the victim of a subsequent attempted murder whom he did not expect to live to tell the tale.

    5.43         It appears from the six decided cases in England and Wales that sections 76 to 79 of the 2003 Act require the Court of Appeal, in considering whether to grant an application to have an acquittal set aside and a retrial ordered, to form its own view not only as to the strength of the new evidence in itself, but also of whether it is likely that the jury at a further trial would convict. The Law Commission's aim of establishing a test which does not require the Court of Appeal to speculate upon the likely outcome of future proceedings does not seem to have been met.

    5.44         We may speculate that a requirement that new evidence be compelling in itself would be even less realistic in Scotland than in England and Wales, owing to the distinctive approach which our law takes to corroboration. It is not hard to imagine a case in which a prosecution which is very strong in almost all respects fails for want of corroboration of a crucial issue. If such corroboration were to become available - perhaps by the emergence of a further complainer whose testimony could provide corroboration under the Moorov doctrine - the state of the evidence, considered as a whole, might well be utterly compelling; however, the evidence which supplies the corroboration may be of minimal probative value when considered in isolation. One might of course wish to limit a new evidence exception to evidence which is compelling in itself, since it is only in such cases that the new evidence might be expected significantly to undermine public confidence in the ability of the judicial system to bring the guilty to justice; however, the English experience suggests that any such limitation is unlikely to be stable. Once the principle is established that acquittals may be reviewed where new evidence shows them to be incorrect, it is hard to maintain as a matter of principle that only certain types of evidence should be considered.

    "Miscarriage of justice"

    5.45         Rather than focusing directly upon the compelling nature of the new evidence, or speculating upon the outcome of a future trial, it would be possible to focus the test upon the effect which the new evidence would have had had it been available at the original trial. Such an approach was considered and rejected by the Law Commission:

    "One . . . 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 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 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."[152]

    5.46         So far as the justification for a retrial following a tainted acquittal is concerned, we agree. We also agree that there is a clear difference between having a second trial in circumstances where the first trial was not a proper trial at all and seeking to disturb the verdict of a proper trial. A new evidence exception is truly an exception to the principle that no-one should be tried twice for the same crime, whereas a retrial following a tainted acquittal gives effect to that principle by insisting that there must at least be a single valid trial. We do not, however, accept that it would necessarily be inappropriate to frame a test by reference to what the outcome of the original trial would have been had the new evidence then been available.

    5.47         First, viewed from the perspective of an appeal court considering an application to set aside the original verdict and grant a retrial, there does not seem to us to be any significant difference between a test which asks what the original jury would have decided had the new evidence been available and one which asks what a hypothetical future jury would decide when faced with the totality of the evidence now available. These are simply two ways of framing the same question: faced with the new, expanded, body of evidence, what would a reasonable jury decide?

    5.48         Second, we should remember that the application of a new evidence exception would involve setting aside a verdict of acquittal delivered following a proper first trial. The Law Commission said that this would be appropriate only where "the effect of the new evidence is such that the jury's verdict (legitimately reached after a proper trial) cannot in the interests of justice be allowed to stand". This is rather euphemistic, or at least raises, without answering, the question of the circumstances in which it would be inappropriate in the interests of justice to allow such a (procedurally) valid verdict to stand. At least in outline, we think that the answer is clear: the setting aside of a procedurally valid verdict could only be in the interests of justice where new evidence provided compelling grounds for believing that the original verdict, while procedurally valid, was wrong in substance; that is, compelling grounds for believing that the acquitted person is truly guilty of the crime of which he was acquitted.

    5.49         Against this background, we consider the suggestion that the test for a new evidence appeal should reflect that which presently applies in a defence appeal against conviction on the basis of new evidence. This was the approach favoured both by the Judges of the High Court of Justiciary and the Faculty of Advocates. The Judges said:

    "The test before the Court would set aside the verdict of the jury and allow a new trial requires to be a high one. We are of the view that broadly the same test should be applied to the Crown as is applied to the defence in new evidence appeals namely: that the new evidence is of such significance that it would be reasonable to conclude that the verdict of the jury reached without the benefit of that evidence, should be regarded as a miscarriage of justice."
    The Faculty of Advocates suggested that the test for new evidence should be the same as that which they suggested for tainted acquittals, namely whether there had been a miscarriage of justice according to the criteria set out by the High Court in Megrahi v HMA.[153]

    5.50         We do not agree with the Faculty's suggestion that the tests for new evidence and tainted acquittals should be the same. The two situations are quite different. Only in the new evidence case is the prosecution seeking to overturn the verdict of a properly conducted, procedurally valid trial. This should, in principle, be subject to a more exacting test than that which should apply once it has been established that the original proceedings were affected by a deliberate attempt to pervert the course of justice.

    5.51         It is appropriate, for the sake of convenience, to quote again the passage from Megrahi to which we have already referred in discussing the test for tainted acquittals:

    " . . . the next question that requires to be addressed is the content of the proposed additional evidence, and whether its significance is such that the fact that it was not heard by the trial court would be regarded as having resulted in a miscarriage of justice . . . We summarise the approach adopted in those cases in the following propositions: (1) The court may allow an appeal against conviction on any ground only if it is satisfied that there has been a miscarriage of justice. (2) In an appeal based on the existence and significance of additional evidence not heard at the trial, the court will quash the conviction if it is satisfied that the original jury, if it had heard the new evidence, would have been bound to acquit. (3) Where the court cannot be satisfied that the jury would have been bound to acquit, it may nevertheless be satisfied that a miscarriage of justice has occurred. (4) Since setting aside the verdict of a jury is no light matter, before the court can hold that there has been a miscarriage of justice it will require to be satisfied that the additional evidence is not merely relevant but also of such a significance that it will be reasonable to conclude that the verdict of the jury, reached in ignorance of its existence, must be regarded as a miscarriage of justice. (5) The decision on the significance of the additional evidence is for the appeal court, which will require to be satisfied that it is important and of such a kind and quality that it was likely that a reasonable jury properly directed would have found it of material assistance in its consideration of a critical issue at the trial. (6) The appeal court will therefore require to be persuaded that the additional evidence is (a) capable of being regarded as credible and reliable by a reasonable jury, and (b) likely to have had a material bearing on, or a material part to play in, the determination by such a jury of a critical issue at the trial." [154]

    5.52         The court in this passage was summarising existing authority, and the leading case remains that of Cameron v HMA.[155] In that case, the High Court suggested that the consequences of a finding of miscarriage of justice should depend upon the degree to which the new evidence demonstrates the innocence of the accused. If the new evidence is such that the court is satisfied that the original jury, had it heard the new evidence, would have been bound to acquit, the court should quash the conviction. If, on the other hand, the appeal court is satisfied merely that the additional evidence is important and reliable evidence which would have been likely to have had a material bearing on, or a material part to play in, the jury's determination of a critical issue at the trial, the court should hold that a conviction returned in ignorance of that evidence represents a miscarriage of justice, set aside the verdict, and authorise a new prosecution.

    5.53         It is superficially tempting to adopt the same approach to a new evidence exception, setting aside the original verdict and allowing a new prosecution wherever the new evidence would have been likely to have had a material bearing on, or a material part to play in, the jury's determination of an issue at the original trial.

    5.54         However, we consider that the straightforward analogy with an appeal against conviction is misleading. The test of whether a conviction returned in the absence of a particular piece of evidence represents a miscarriage of justice does not amount to a test of whether the original verdict was substantially correct; rather, it represents, first, a recognition that the accused is entitled to the benefit of any doubt which might arise at his trial and, second, a recognition that the determination of guilt is for the jury. Simply put, it is important to give the accused (or the appellant) the benefit of any doubt that there may be about whether new evidence would have caused the jury to return a different verdict.[156] The same benefit of the doubt cannot be extended to the prosecutor who seeks to disturb a verdict of acquittal.[157]

    5.55         While it would thus be inappropriate to apply a test of whether the new evidence is "of such a significance that it will be reasonable to conclude that the verdict of the jury, reached in ignorance of its existence, must be regarded as a miscarriage of justice", there may be merit in limiting the test to the other type of situation identified in Cameron and Megrahi. On such a test, the Lord Advocate's application would be granted only if the High Court could be satisfied that if a reasonable jury were to have heard the evidence as it emerged at the original trial, together with the new evidence, that jury would have been more or less bound to convict.

    Conclusions

    5.56         The preceding survey of the possible tests, combined with the English experience of a "compelling evidence" test, strongly suggests that any test which recognises the burden of proof in criminal cases and the value of finality in the verdicts of properly conducted proceedings will require the court hearing the application to base its decision at least partly upon its own view of whether the evidence now shows the accused to be guilty. It may be just about possible in principle to frame a test which refers only to the significance of the new evidence, asking whether this is compelling in itself; however, the English experience suggests that the courts are unlikely to be attracted to this somewhat artificial approach when faced with an application in which the new evidence, while not compelling in itself, does add to the existing evidence to form a compelling case of guilt. There seems, in practice, no escape from having the court pronounce a view as to whether or not the new evidence establishes the accused's guilt.

    5.57         Any test which focused upon new evidence would be highly artificial. The court would require to proceed on the assumption that the new evidence, had it been available, would not have affected the way in which the other evidence came out at the first trial (or, the same thing, that at a new trial at which the new evidence is led, the other evidence would emerge exactly as it did at the original trial). The unrealistic nature of any such test was noted by the Appeal Court in Fraser v HMA, where Lord Osborne observed that:

    "while the assessment of the significance of the additional evidence not heard at the trial must be conducted in the context of the whole evidence laid before the trial court, in my opinion, the very existence of that additional evidence inevitably means that a new evidential situation has been created, which is bound to render the tactics actually adopted at the trial, in the light of the earlier different evidential situation, obsolete and irrelevant."[158]

    5.58         This point is more significant in the context of a prosecution application than it was in the Fraser appeal, since the consequence of the introduction of new prosecution evidence might be a complete change in the basis upon which the defence is conducted, perhaps including the evidence which the defence would seek to lead. While the new evidence may utterly undermine the original defence, this does not mean that there might not be another valid, or at least arguable, defence which was not relied upon at the original trial. For example, a person accused of murder might, in fact, have a sound defence of self defence. However, aware that this defence would be hard to establish, he may have preferred to construct a false alibi. On the basis of the evidence led at the first trial, new evidence proving the falsity of the alibi would be compelling evidence of guilt; but in any new trial, the defence, aware of the existence of this evidence, would be conducted on a wholly different basis.

    5.59         In addition, it should be borne in mind that the court considering the application will not have heard the witnesses at the original trial. That court will necessarily be restricted to considering a transcript of the earlier evidence. The strength of a body of evidence depends not only upon what was said by the witnesses, but also upon the credibility of those witnesses. A court which does not hear the witnesses is in a poor position to assess their credibility, and would have great difficulty in concluding on the basis of a transcript of the original evidence that any reasonable jury hearing that evidence together with the new evidence would have been bound to convict. The highest test that could in practice be satisfied would be one which required the court to conclude that had the new evidence been heard by the original jury, that jury would have been highly likely to have convicted the accused.

    A proposed test

    5.60         We do not think that it could ever be appropriate to set aside a procedurally valid verdict of acquittal except where the new evidence both substantially strengthens the case against the acquitted person and, taken together with the evidence as heard at the first trial, makes it highly likely that the accused was in fact guilty.

    5.61         Each element is essential: the first, to prevent what is in substance an appeal against the jury's verdict in a case where there is truly little new evidence but where an apparently strong original case led to a surprising acquittal; the second since the only justification for setting aside a verdict obtained in a proper trial is that the verdict can be shown, with some high degree of confidence, to be incorrect. We do not ultimately share the Law Commission's reasons for rejecting each of these elements. Neither, considered in isolation, is adequate; the two together work as well as any test can. We recognise that the test of substantially strengthening the case against the accused will be more easily met where the original case was weak; however, the test of whether the evidence, considered as a totality, is compelling will be correspondingly harder to meet. We accept that the test requires the court hearing the application to form its own view as to the merits of the case against the accused. We accept too that this is a defect; but better this than allowing a wider range of acquittals to be opened to challenge.

    5.62         We recommend:

    33. An application on the basis of new evidence to have an acquittal set aside and authority granted for a new prosecution should be granted only where the court is satisfied –
    (a) that the new evidence substantially strengthens the case against the accused; and
    (b) that, had a reasonable jury heard the evidence at the original trial, together with the new evidence, it is highly likely that the accused would have been convicted.
    (Draft Bill, section 8(7)(a), (c))
    The interests of justice

    5.63         Respondents to the Discussion Paper supported our suggestion that whatever test the court was required to apply in considering the application of a new evidence exception, that court should have a discretion not to authorise a retrial where it appears to the court not to be in the interests to justice to do so.

    5.64         There are a number of factors which might render a retrial contrary to the interests of justice. The question of whether it will be possible to have a fair trial given the existence of prejudicial publicity will be one, although the court might well feel it more appropriate to consider this in the context of a plea of oppression raised at the outset of any subsequent proceedings. Other factors can readily be imagined, relating to the passage of time and the unavailability of defence witnesses. We do not think it necessary, or particularly helpful, to seek to constrain the court's freedom to deny the Crown the opportunity of a second prosecution where it would, in the court's view, be contrary to the interests of justice, provided only that the courts should not exercise this discretion so as to reintroduce the strict prohibition on double jeopardy by the back door.[159]

    5.65         We recommend:

    34. The court considering an application to authorise a retrial on the basis of new evidence should decline to grant that application where it considers that to grant the application would be contrary to the interests of justice.
    (Draft Bill, section 8(8))
    Reporting restrictions

    5.66         A significant number of cases have dealt with the question of the effect of potentially prejudicial publicity upon the fairness of a criminal trial. The incorporation into domestic law of the ECHR has had an impact on the development of the law, inasmuch as it is no longer possible, as it may once have been, to balance the interest of the accused in having a fair trial against the public interest in the prosecution of offences.[160] The right of the accused to a fair trial by an independent and impartial tribunal is unqualified, and cannot be subordinated to the public interest in the detection and suppression of crime.[161] Subject to this qualification, the test in Scotland remains that set out by the High Court of Justiciary in Stuurmann v HMA,[162] namely that the High Court should intervene to prevent the Lord Advocate from proceeding upon an particular indictment only where to do so would amount to oppression, and that oppression will arise only where the risk of prejudice is so grave that no direction of the trial judge could reasonably be expected to remove it.

    5.67         This is a high test to meet, and reported cases in which potentially prejudicial publicity has led to a successful plea of oppression are exceedingly rare. One feature which distinguishes the most serious cases is the proximity in time of the prejudicial reporting to the trial, since one of the factors which has been held to dilute the potentially prejudicial effect of such reports is the lapse of time between the reports and the trial and the consequent fading of jurors' recollections. Prejudicial publicity immediately prior to the trial has led to the desertion of trial diets pro loco et tempore,[163] and prejudicial publicity during trials has led to permanent stays of proceedings in a number of reported cases in England and Wales.[164]

    5.68         In relation to a retrial on the basis of new evidence, there would clearly be a substantial risk of prejudice were the jury at the second trial to be aware of the details of the process by which the case had come before them. A jury member who was aware that three judges of the High Court had concluded that the totality of the evidence which they expected to be led would effectively compel a reasonable jury to deliver a guilty verdict might well be influenced by that knowledge and fail to reach her own view as to the effect of the evidence which was actually led. In such a case there would be a real risk that the jury would be so prejudiced in its consideration of the evidence at the second trial that the second trial could not be a fair one. This risk would be particularly marked in any case in which the evidence at the second trial failed, for whatever reason, to come up to the level expected when the application for a new trial was granted.[165]

    5.69         If the jury were to become aware of the procedural history of the case during the course of the second trial, then the prejudicial effect might be argued to be more significant than in a case where there was prejudicial publicity during a trial, since the opinion of the three judges who considered the application would be more authoritative and immediate than any press report. But there is no reason why a second jury should be aware of the procedural history of the case. It is certainly possible, and perhaps even likely, that such a jury will become aware that they are sitting on a retrial - for instance, one or other party may wish to cross-examine a witness on the basis of a change in testimony between the first and second trials. But retrials are competent in a range of circumstances. It is hard to imagine a case in which the jury would require to be told, or would accidentally be made aware, that the retrial was on the basis of new evidence and that the procedure for authorising a retrial on the basis of such evidence involved three judges of the High Court taking a view as to the compelling nature of the evidence against the accused. The only route by which such detailed knowledge of the procedure leading up to the retrial is likely to come to the jury is via media reporting. In the Discussion Paper we proposed that the court should have power to postpone reporting of any decision to grant authority for a new trial on the basis of new evidence, in order to minimise the risk of any such potentially prejudicial knowledge coming into the hands of the jury. There was widespread support for this proposal. We also recognise that a risk of prejudice to the new trial might arise from reporting, during the application proceedings, of details of the original trial. For this reason, we think that it would be appropriate for the application and any new trial to be treated as active proceedings in terms of the Contempt of Court Act 1981 from the date on which the application is made, making it a strict liability contempt of court to publish any material which creates a substantial risk of prejudice to those proceedings.

    5.70         We propose:

    35. The court to which an application is made for a retrial based upon new evidence should have power to make an order limiting publication of reports of the application and its result.
    (Draft Bill, section 11)
    Retrospectivity

    5.71         The introduction of a retrospective exception to the rule against double jeopardy would not in itself criminalise any conduct committed before the law came into force. If the penalties for the offence in question had been increased, no doubt appropriate provision could be made to prevent a more severe penalty being imposed than was competent at the time of the original offence.[166] But to make the provision retrospective would have a novel and irreversible effect upon those who had been acquitted of crimes. At present any person acquitted of a crime can be certain that that part of his life is past. He can carry on secure in the knowledge that that matter cannot be raised again in the criminal courts. As we have noted above, a person who has been tried for an offence and acquitted has a right, recognized and enforced by the courts, not to be tried again for the same offence.

    5.72         The Law Commission considered retrospectivity on the basis that the change was essentially procedural.  They said:

    "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."[167]

    5.73         We are not sure that we follow the logic of this argument, not least because it appears to beg the question which the earlier acquittal has answered. The starting point is that any accused person is and remains innocent of an offence until he is proved to be guilty. When an offence has been committed, every citizen is liable to be charged with having committed it, and to be tried. If one citizen is acquitted, it is open to the authorities to charge another with the offence, and to try him. But at present, every citizen has the expectation that he will be tried only once for any offence, and those citizens who have been tried already for a particular offence have not simply an expectation; they have a right not to be tried again for that offence. It is the liability to the stress and risk of criminal proceedings which constitutes the "greater liability" to which previously acquitted persons are exposed if the provision is made retrospective.

    5.74         It is certainly possible to draw an analogy between the introduction of a new evidence exception and purely procedural changes.  The Law Commission referred to changes in the rules of admissibility of evidence.  The perpetrator of a crime who is justly confident of escaping justice on the basis that he knows the only evidence against him to be inadmissible is not unfairly prejudiced by a change in the law which allows the evidence of his conduct - a crime at the time of its commission - to form the basis of a prosecution. But the right not to be tried twice for the same offence is of a different, and superior, order. It reflects not simply a current procedural practice, but an expression of a long-standing public interest in the finality of court judgments, in the protection of the citizen from repeated prosecutions on the same subject, and, more generally, in the recognition of the status of the individual against the state. The person who believes that the evidence against him is inadmissible is still liable to be tried. In the course of that trial his opinion of the inadmissibility of the evidence may or may not be found to be correct. But the person who has been tried already is in a wholly different position. He can prevent the proceedings from taking place at all.

    5.75         In our view, that right of an acquitted person not to be tried again is an actual right which is recognized and enforced by the courts. It enables an acquitted person to pursue his private life untroubled by the prospect of the repetition of the court proceedings to which he has been subjected in the public interest. It is for consideration whether interference with that right might constitute a breach of an individual's rights under Article 8 of the Convention.[168] We are not aware of any case in which such a claim has been brought, but we are conscious that "private life" is a term which is interpreted widely and which covers the physical and psychological or moral integrity of a person.[169] The easy answer is that Article 8 is not engaged by a decision to make any exception to the rule retrospective. But the right not to be tried again is, as noted above, an actual right, enforceable in the courts, and currently enjoyed by every citizen who has been acquitted of an offence. If Article 8 were engaged, then other questions would arise. It is open to the state to interfere with Article 8 rights, if that is in the public interest and otherwise in accordance with the requirements of Article 8(2). And provisions which, for the future, changed the basis of the criminal liability of all citizens are likely to be considered to be a product of the balancing of different interests which is envisaged by Article 8(2).

    5.76         But if the position of persons already acquitted is altered then they are being deprived of a right which they currently enjoy. And any interference with rights which are protected by the Convention must be proportionate to the public interest to be served. Before removing the right not to be tried again, we might expect the state to be able to demonstrate some counterbalancing public interest in securing the conviction of probable criminals who, under the present law, are able to avoid justice because of the operation of the rule against double jeopardy. As we have noted more than once in this Report, there is no evidence of even one person in relation to whom such a public interest could be demonstrated.

    5.77         Nevertheless, we have - for obvious reasons - seen no case in which the point has been taken, and we do not consider, absent any authority on the matter, that we could definitively say that the retrospective application of a new evidence exception would engage Article 8 of the Convention.

    5.78         We are of course conscious that the question as to compatibility with Article 8 might be said to have been settled by the seventh Protocol, and in particular by Article 4 of that Protocol.[170] But that Article simply makes it competent to introduce provision as to the retrial of persons if new evidence becomes available. It does not require any state to introduce such provision; still less does it say anything about the transition from a position in which such retrials are not competent to a position in which they are.

    5.79         Leaving aside any question of the European Convention, there is a question as to whether this alteration to the position of acquitted persons (to their detriment) is justifiable on general public policy grounds. We have seen no evidence of any reason to make such a provision retrospective. None of our respondents has mentioned any case in which retrospection would potentially open the possibility of retrying a person against whom new evidence has become available. Nor, as we have discussed, does there appear to be any prospect of technological advances assisting in this matter, since the physical evidence and productions to which such developments might be related will have been destroyed as a matter of routine prosecution practice.

    5.80         On the other hand, as the Director of Public Prosecutions pointed out in his response:

    "A rule which addressed the potential injustice of an unmeritorious acquittal for future cases only might be thought of as somewhat artificial. The public is unlikely to understand why someone seemingly linked to a serious and perhaps notorious crime by new and apparently compelling evidence cannot be brought to justice merely because their wrongdoing occurred before a certain date."

    5.81         It might be said that this observation misses the point, however.  Prior to the introduction of an exception, the state will have one opportunity, and one opportunity only, of convicting an accused citizen in a criminal trial. The result of the judges' action in promoting the rule against double jeopardy is that, following such a trial, the acquitted person is entitled to continue with his life free from the threat of further proceedings.  No such protection exists in the case of, say, evidential rules or extradition arrangements. 

    5.82         The core of the principled objection to a retrospective exception is that there is a difference between saying to a person who may in future be accused of a crime that the verdict at his trial may not be final, and saying to a person who has duly submitted to the criminal process that the verdict that he believed, correctly, to be final is only provisional after all. 

    5.83         Furthermore, this reneging upon the previous understanding of the nature of acquittals would affect all of those acquitted of offences of a type to which the exception might apply.  All such persons, whether in fact innocent or guilty, would have had their unconditional verdicts rendered provisional at a stroke.

    5.84         The Director of Public Prosecutions may be correct in saying that the public would be unlikely to understand it, but as a matter of principle we think that this argument has merit.  We consider that in order to justify the retrospective application of a new evidence exception, one would need to be satisfied that the benefit of doing so would be sufficient to outweigh the disadvantage of departing from the explicit representation of finality which has been made to those already tried.

    5.85         None of the respondents to the Discussion Paper was able to point to any existing case in which a new evidence exception could be applied.  COPFS (who were strongly in favour of retrospectivity) suggested that one reason for this was that in the absence of such an exception there was no incentive to look for such cases.  This may well be so.  Nevertheless, the absence of any evidence of existing cases in which the retrospective application of an exception would enable the correction of allegedly erroneous acquittals strikes us as telling.

    5.86         We have also had regard to what COPFS has told us about the treatment of productions[171] following an acquittal.  At present, while the case papers are retained, productions are routinely disposed of after the end of a trial which results in acquittal.  If a new evidence exception were introduced for future cases, we might expect this practice to change; but this practice, perfectly appropriate in view of the current law, must in our view severely restrict the practical benefit to be gained from making any new evidence exception retrospective in application.

    5.87         In summary, we consider that while it would probably be competent for the Scottish Parliament to legislate to give a new evidence exception retrospective effect, this should only be done if the Parliament is satisfied that the practical benefits of retrospective application in terms of the correction of erroneous acquittals in serious cases are sufficient to outweigh the general detriment to all those previously acquitted in rendering their hitherto final verdicts open to challenge.  On the basis of the evidence provided in response to our Discussion Paper, we do not consider that this test is met.

    5.88         We propose:

    36. Any exception to the rule against double jeopardy on the basis of new evidence should apply only to cases originally determined after the coming into force of the exception.
    (Draft Bill, section 8(1))
     
    Part 6 Summary of recommendations
  8. There should continue to be a general rule against double jeopardy.
  9. The general rule against double jeopardy should be reformed and restated in statute.
  10. A second prosecution should be prohibited where a person has previously been convicted or acquitted of an offence and:
  11. (a) the second indictment or complaint charges an offence of which it would have been competent to convict the accused on the first indictment or complaint (so, for instance, an earlier verdict on a trial for murder will bar subsequent prosecution for attempted murder, culpable homicide, assault etc.); or
    (b) the second indictment or complaint charges an offence which is an aggravated form of the offence charged on the first occasion (so, for instance, a previous conviction or acquittal of assault will bar a subsequent prosecution for assault to severe injury).
    (Draft Bill, section 1(1))
  12. Beyond recognising a rule against trying a person again for an offence of which he has already been convicted or acquitted, Scots law should recognise a broader principle that a person should not be tried again in relation to the same acts which gave rise to that prosecution.
  13. (Draft Bill, section 2)
  14. It should continue to be possible to prosecute a person for murder or culpable homicide where that person has previously been tried and convicted, prior to the victim's death, for an offence involving the assault which is alleged to have led to the victim's death.
  15. (Draft Bill, section 3(4))
  16. It should, however, no longer be possible to prosecute a person for murder or culpable homicide where that person has previously been acquitted of an offence involving the assault which is alleged to have led to the victim's death.
  17. (Draft Bill, section 3(3))
  18. It should be possible to prosecute a person for culpable homicide or for a statutory offence of causing death where that person has previously been tried and convicted, prior to the victim's death, for an offence relating to the act or omission which is alleged to have led to the victim's death; such a prosecution should, however, be barred where the outcome of the first trial was acquittal.
  19. (Draft Bill, section 3(3), (4))
  20. After the verdict at any trial for homicide following an earlier conviction for an offence relating to the act which is alleged to have caused the victim's death, the accused should be able to make a motion that the earlier verdict be quashed in light of the verdict at the homicide trial.
  21. (Draft Bill, section 3(4)(b))
  22. The rule against double jeopardy should apply to bar a subsequent prosecution where there has been a finding of guilt, or the acceptance by the prosecutor of a guilty plea, regardless of whether sentence has been imposed.
  23. (Draft Bill, section 1(4))
  24. A verdict of conviction or acquittal by a foreign court should bar a subsequent prosecution in Scotland in respect of the same acts which gave rise to the first prosecution; but the Scottish court should be permitted to disregard an acquittal or conviction in a foreign jurisdiction where it is satisfied that it is in the interests of justice to do so.
  25. (Draft Bill, section 2(4), (5)(a))
  26. In considering whether to disregard a foreign verdict, the court should have regard, among other considerations, to –
  27. (a) the extent to which the acts constituting the offence charged are alleged to have taken place in Scotland;
    (b) the relative seriousness with which the law of Scotland and that of the foreign jurisdiction view the conduct which forms the subject matter of the instant charge;
    (c) whether it appears that the foreign proceedings were held for the purpose of shielding the accused from criminal responsibility, were not conducted independently or impartially, or were conducted in a manner inconsistent with an intent to bring the perpetrator of the alleged offence to justice.
    (Draft Bill, section 2(6))
  28. In any case where the Crown would propose to argue in response to a plea which might be raised in relation to double jeopardy that the original proceedings were a nullity, the approval of the High Court should be required before proceedings may be brought.
  29. (Draft Bill, section 10)
  30. It should be possible to retry an acquitted person where that person's acquittal was tainted by an offence against the administration of justice in relation to the original trial.
  31. (Draft Bill, section 4)
  32. It should not be necessary, in order to justify such a retrial, to show any involvement of the acquitted person in the offence against the administration of justice.
  33. (Draft Bill, section 4(3)(a))
  34. It should continue to be competent to prosecute a person for perjury where there is evidence that he committed perjury in giving evidence on his own behalf in prior criminal proceedings.
  35. Before authorising a new prosecution on the basis of a tainted acquittal, it must be established, on the balance of probabilities, that an offence against the administration of justice was committed in relation to the original trial.
  36. (Draft Bill, section 4(3)(a))
  37. The commission of such an offence may be proved either by the production of a relevant extract conviction, or by evidence led by the prosecutor.
  38. (Draft Bill, section 4(3)(a))
  39. An acquittal should be regarded as tainted where, but for the commission of an offence against the administration of justice involving interference with a witness, the witness in question would have either –
  40. (i) given evidence which was capable of being regarded as credible and reliable by a reasonable jury and which would, if accepted by the jury, have had a material bearing on, or a material part to play in, the determination by them of a critical issue at the trial; or
    (ii) refrained from giving perjured evidence which, assuming that it was accepted by the jury, was likely to have had a material bearing on, or a material part to play in, the determination by them of a critical issue at the trial.
    (Draft Bill, section 4(14))
  41. Where it is established that there has been interference with a juror or with the judge in the proceedings in which the accused was acquitted, that acquittal should be regarded as tainted unless the court hearing the application is satisfied that the interference has had no effect on the proceedings.
  42. (Draft Bill, section 4(11), (12))
  43. Any application to have an acquittal set aside as tainted and authority granted for a new prosecution should be made to a quorum of three judges of the High Court.
  44. (Draft Bill, section 4(6))
  45. The court considering an application under the tainted acquittal procedure should refuse that application where it appears to the court that to retry the accused would be contrary to the interests of justice.
  46. (Draft Bill, section 4(15))
  47. The court hearing an application under the tainted acquittal procedure should have the power to impose restrictions upon the reporting of the application, with a view to preventing the publication of such reports from influencing the jury in any subsequent trial.
  48. (Draft Bill, section 11)
  49. The application of the tainted acquittal procedure should not be restricted by reference to the court (solemn or summary) before which the original proceedings took place.
  50. (Draft Bill, section 4(1))
  51. The application of the tainted acquittal procedure should not be limited by reference to the date of commission of either the original offence, the offence against the course of justice which triggers the application of the procedure, or to the date of the allegedly tainted acquittal.
  52. (Draft Bill, section 4(2))
  53. It should be possible, with the authority of the High Court, to reprosecute a person who confesses to having committed an offence of which that person has previously been tried and acquitted.
  54. (Draft Bill, section 6)
  55. Application for authority to reprosecute should be made by the Lord Advocate to a quorum of three judges of the High Court.
  56. (Draft Bill, section 6(3), (6))
  57. The High Court should grant the application only if satisfied-
  58. (a) that the acquitted person has admitted having committed the crime to which the acquittal relates;
    (b) that the admission is credible and reliable; and
    (c) that it would not be contrary to the interests of justice to reprosecute the acquitted person.
    (Draft Bill, section 6(7), (8))
  59. The application of the exception for post-acquittal confessions should not be limited by reference to the date of commission of the offence of which the accused was originally acquitted or by the date of the verdict of acquittal.
  60. (Draft Bill, section 6(2))
  61. A new evidence exception should apply only in relation to the offences of murder and rape.
  62. (Draft Bill, section 8(9))
  63. Scottish Ministers should be able to add further offences to the list by way of affirmative order.
  64. (Draft Bill, section 8(10))
  65. For the purpose of a new evidence exception, evidence should be regarded as "new" only if it was not, and could not with the exercise of reasonable diligence have been, available at the original trial.
  66. (Draft Bill, section 8(7)(b))
  67. Evidence which was available to the prosecution (or could, with ordinary diligence, have been available) at the first trial but which was then inadmissible should not be regarded as "new" by virtue of any subsequent change in the rules relating to the admissibility of evidence.
  68. An application on the basis of new evidence to have an acquittal set aside and authority granted for a new prosecution, should be granted only where the court is satisfied –
  69. (a) that the new evidence substantially strengthens the case against the accused; and
    (b) that, had a reasonable jury heard the evidence at the original trial, together with the new evidence, it is highly likely that the accused would have been convicted.
    (Draft Bill, section 8(7)(a), (c))
  70. The court considering an application to authorise a retrial on the basis of new evidence should decline to grant that application where it considers that to grant the application would be contrary to the interests of justice.
  71. (Draft Bill, section 8(8))
  72. The court to which an application is made for a retrial based upon new evidence should have power to make an order limiting publication of reports of the application and its result.
  73. (Draft Bill, section 11)
  74. Any exception to the rule against double jeopardy on the basis of new evidence should apply only to cases originally determined after the coming into force of the exception.
  75. (Draft Bill, section 8(1))
    Appendix A
    Double Jeopardy (Scotland) Bill
    [DRAFT]
    CONTENTS
    Section
  76. Rule against double jeopardy
  77. Plea in bar of trial
  78. Eventual death of injured person
  79. Tainted acquittals
  80. Further provision as regards prosecution by virtue of section 4
  81. Admission subsequent to acquittal
  82. Further provision as regard prosecution by virtue of section 6
  83. New evidence
  84. Further provision as regards prosecution by virtue of section 8
  85. Nullity of proceedings on previous indictment or complaint
  86. Amendment of Schedule 1 to the Contempt of Court Act 1981
  87. Short title, interpretation and commencement
  88. Double Jeopardy (Scotland) Bill
    [DRAFT]
    An Act of the Scottish Parliament to make provision as to the circumstances in which a person convicted or acquitted of an offence may be prosecuted anew; and for connected purposes.
    1 Rule against double jeopardy
    (1) It is not competent to charge a person who, whether on indictment or complaint (the "original" indictment or complaint), has been convicted or acquitted of an offence—
    (a) with an offence of which it would have been competent to convict the person on the original indictment or complaint, or
    (b) with an offence which—
    (i) arises out of the same, or largely the same, acts or omissions as gave rise to the original indictment or complaint, and
    (ii) is an aggravated way of committing the offence of which the person was convicted or acquitted.
    (2) Whether the conviction or acquittal was before or after the coming into force of this section is, for the purposes of the section, immaterial.
    (3) Subsection (1) is subject to sections 4, 6 and 8 and is without prejudice to sections 118(1)(c) (disposal of appeals), 119 (provision where High Court authorises new prosecution), 183(1)(d) (stated case: disposal of appeal) and 185 (authorisation of new prosecution) of the 1995 Act.
    (4) In this Act, reference to a person being convicted of an offence is—
    (a) to the person being found guilty of the offence, or
    (b) to the prosecutor accepting the person's plea of guilty to the offence,
    in either case whether or not sentence is passed.
    NOTE
    Section 1 restates the core rule against double jeopardy (Recommendation 3, para 2.18). Subsection (1) declares it to be incompetent to charge a person who has previously been convicted or acquitted of an offence with an offence of which that person could have been convicted upon the original indictment or complaint, or with an aggravated version of that offence. The section does not prevent a person from being tried for murder or culpable homicide where the victim dies after that person's conviction of assault, since murder and culpable homicide are not aggravated ways of committing assault but separate crimes; such prosecutions are regulated by section 3.
    An issue relating to the competency of an indictment is a "preliminary plea" in terms of section 79(2)(a)(i) of the Criminal Procedure (Scotland) Act 1995 (the "1995 Act"), which must be raised by the accused prior to the first diet or preliminary hearing. Under summary procedure, section 144(4) of that Act requires any objection to the competency of a summary complaint to be stated at the first diet before the accused pleads to the charge. Whether under solemn or summary procedure, a decision as to the competency of a charge is open to appeal by either party with the leave of the court of first instance (1995 Act, s 74(1) (proceedings on indictment), s 174 (summary proceedings)).
    Subsection (2) provides that section 1 shall apply to all cases, regardless of whether the original verdict was delivered before or after the coming into force of the section.
    Subsection (3) makes it clear that section 1 does not bar a further prosecution where this is authorised under section 4, 6 or 8 of the Bill, or under the provisions of the 1995 Act which allow the High Court to authorise a new prosecution following an appeal against conviction on indictment or an appeal by way of stated case against the verdict in a summary prosecution.
    Subsection (4) defines "conviction" for the purposes of the Bill. This definition settles the question of whether a sentence must be passed before the rule against double jeopardy may operate (Recommendation 9, para 2.62-2.63), making it clear that double jeopardy protection will apply in any case where a verdict has been delivered or a guilty plea accepted, regardless of whether sentence has been passed.
    2 Plea in bar of trial
    (1) A person charged with an offence—
    (a) whether on indictment or complaint, but
    (b) other than by virtue of a section mentioned in section 1(3),
    may aver, as a plea in bar of trial, that the offence arises out of the same, or largely the same, acts or omissions as have already given rise to the person being tried for, and convicted or acquitted of, an offence (the "original offence").
    (2) Whether the conviction or acquittal was before or after the coming into force of this section is, for the purposes of the section, immaterial.
    (3) If the court is satisfied, on a balance of probabilities, as to the truth of the person's averment, the plea is to be sustained unless the prosecutor persuades the court that there is some special reason why the case should proceed to trial (as for example, but without prejudice to the generality of this subsection, where trials were separated on the application of, or with the consent of, the person).
    (4) Subsections (1) to (3) apply irrespective of where the person was tried; but this subsection is subject to subsection (5).
    (5) Where the person was tried outwith the United Kingdom the court may disregard a conviction or acquittal if—
    (a) it determines that it is in the interests of justice to do so, and
    (b) to permit the case to proceed to trial would not be inconsistent with the obligations of the United Kingdom under Article 54 of the Schengen Convention (that is to say, of the Convention of 19th June 1990 implementing the Schengen Agreement of 14th June 1985).
    (6) In making a determination in pursuance of subsection (5)(a), the court is in particular to have regard to—
    (a) whether the purpose of bringing the person to trial in the foreign country appears to have been to assist the person to evade justice,
    (b) whether the proceedings in the foreign country appear to have been conducted—
    (i) independently and impartially, and
    (ii) in a manner consistent with dealing justly with the person,
    (c) whether such sentence (or other disposal) as might be imposed in the foreign country for an offence of the kind for which the person has been acquitted or convicted is commensurate with any that might be imposed for an offence of that kind in Scotland, and
    (d) the extent to which the acts or omissions can be considered to have occurred in, respectively—
    (i) Scotland,
    (ii) the foreign country.
    NOTE
    Section 2 enacts the broader principle against multiple trials in relation to the same acts (Recommendation 4, paras 2.19-2.27) by allowing a person accused of a crime to plead in bar of trial that the crime charged against him relates to the same acts, or substantially the same acts, as a crime of which he has already been convicted or acquitted (subsection (1)). If the court is satisfied, on the balance of probabilities, that the prosecution does arise from substantially the same acts as an earlier trial, it must sustain the plea in bar of trial unless satisfied that there is some special reason why the case should proceed to trial. (Subsection (3)). The section gives, as an example of such a special reason, the case in which trials were separated on the application of, or with the consent of, the person against whom the charge is brought. (For a discussion of what might constitute a special reason, see paras 2.32-2.35).
    This section also makes provision for the treatment of foreign verdicts (Recommendations 10 and 11, paras 2.64-2.74). The general rule is that, for the purposes of the plea in bar, it does not matter whether the original trial took place in Scotland or elsewhere. However, if the person was originally tried outwith the United Kingdom, the court may disregard a conviction or acquittal where it determines that it would be in the interests of justice to do so (subsection (5)(a)). In determining whether it is in the interests of justice to disregard a foreign verdict for the purposes of section 3, the court is required to have regard to the factors listed in subsection (6). The court is prevented from disregarding a non-UK verdict where trying the accused would be inconsistent with the UK's obligations under Article 54 of the Schengen Convention (see para 2.36); that is, where a charge relating to the same acts has been finally determined in another State to which Article 54 of that Convention applies (that is, an EU Member State, Iceland or Norway).
    3 Eventual death of injured person
    (1) This section applies where—
    (a) a person ("A") sustains physical injuries,
    (b) another person ("B") is, whether on indictment or complaint, acquitted or convicted of an offence ("offence Y") which comprises the infliction of the injuries, and
    (c) after the acquittal or conviction A dies, ostensibly from the injuries.
    (2) Whether the conviction or acquittal was before or after the coming into force of this section is, for the purposes of the section, immaterial.
    (3) If B was acquitted of offence Y (and was not then convicted of a different offence, "offence Z", which comprised the infliction of the injuries) it is not competent to charge B with—
    (a) the murder of A,
    (b) culpable homicide as respects A, or
    (c) any other offence comprising causing A's death.
    (4) If B was convicted of offence Y (or of offence Z), then—
    (a) for the purposes of sections 1 and 2 the offences mentioned in paragraphs (a) to (c) of subsection (3) are not to be treated as offences arising out of the same, or largely the same, acts or omissions as the offence of which B was convicted, but
    (b) on B being acquitted or convicted of any of the offences mentioned in those paragraphs, the court may, on the motion of B and after hearing the parties on that motion, quash B's conviction of offence Y (or offence Z) where satisfied that it is appropriate to do so.
    (5) A party may appeal to the High Court against the granting or refusing of a motion under subsection (4)(b).
    NOTE
    Section 3 governs cases in which a person is tried and convicted or acquitted of an offence involving the infliction of physical injuries and the victim subsequently dies, apparently from the injuries (subsection (1)). It gives effect to Recommendations 5 to 8 (see paras 2.37-2.61). Subsection (3) provides that where the person was acquitted of the original offence, and was not convicted of any other offence comprising the causing of the injuries, it shall not be competent to charge that person with murder, culpable homicide or any other offence comprising causing the death of the victim of the original offence. Subsection (4)(a) has the effect of allowing a prosecution for murder, culpable homicide, or any other offence comprising the causing of the victim's death in any case in which the first trial resulted in the conviction of the accused for an offence which comprised the infliction of the injuries.
    Subsection (4)(b) gives the trial court, in any case in which the accused is then acquitted or convicted of an offence comprising causing the death of the victim, the power, on the motion of the accused, to quash the earlier conviction for the lesser offence where satisfied that it is appropriate to do so. Subsection (5) grants a right of appeal to both defence and prosecution against the granting or refusing of such a motion.
    4 Tainted acquittals
    (1) A person who, whether on indictment or complaint (the "original" indictment or complaint), has been acquitted of an offence (the "original offence") may, provided that the conditions mentioned in subsection (3) are satisfied, be charged with, and prosecuted anew for—
    (a) the original offence, or
    (b) an offence arising out of the same, or largely the same, acts or omissions as gave rise to the original offence.
    (2) Whether the acquittal was before or after the coming into force of this section is, for the purposes of the section, immaterial.
    (3) The conditions are—
    (a) either—
    (i) that the acquitted person or some other person has (or the acquitted person and some other person have) been convicted of an offence against the course of justice, being an offence in connection with proceedings on the original indictment or complaint, or
    (ii) that on the application of the Lord Advocate the High Court has concluded on a balance of probabilities that the acquitted person or some other person has (or the acquitted person and some other person have) committed such an offence against the course of justice, and
    (b) that on the application of the Lord Advocate the High Court has—
    (i) set aside the acquittal, and
    (ii) granted authority to bring, by virtue of this section, a new prosecution.
    (4) On making an application under subsection (3), the Lord Advocate is to send a copy of that application to the acquitted person.
    (5) The acquitted person is entitled to appear or to be represented at any hearing of the application.
    (6) For the purpose of—
    (a) hearing and coming to a conclusion on any application under subsection (3)(a)(ii), or
    (b) hearing and determining any application under subsection (3)(b),
    three of the Lords Commissioners of Justiciary are a quorum of the Court (the application being determined by majority vote of those sitting).
    (7) The decision of the Court on the application is final.
    (8) Subsection (7) is without prejudice to any power of those sitting to remit the application to a differently constituted sitting of the Court (as for example to the whole Court sitting together).
    (9) The Court may appoint counsel to act as amicus curiae at the hearing in question.
    (10) Subsections (11) and (12) apply in a case where (or as the case may be where the Court, in coming to a conclusion under subsection (3)(a)(ii), is satisfied on a balance of probabilities that) the offence against the course of justice consisted of or included interference with a juror or with the trial judge.
    (11) An acquittal is to be set aside under subsection (3)(b)(i) if the Court is unable to conclude that the interference had no effect on the outcome of the proceedings on the original indictment or complaint.
    (12) But it is not to be so set aside if in the course of the trial, the interference (being interference with a juror and not with the trial judge) became known to the trial judge, who then allowed the trial to proceed to its conclusion.
    (13) Subsection (14) applies in a case other than is mentioned in subsection (10).
    (14) An acquittal is not to be set aside under subsection (3)(b)(i) unless the Court is satisfied on a balance of probabilities—
    (a) that the offence led—
    (i) to the withholding of evidence which, had it been given, would have been, or
    (ii) to the giving of false evidence which was,
    evidence capable of being regarded as credible and reliable by a reasonable jury, and
    (b) that the witholding, or as the case may be the giving, of the evidence was likely to have had a material effect on the outcome of the proceedings on the original indictment or complaint.
    (15) And an acquittal is not to be set aside under subsection (3)(b)(i), whether by virtue of subsections (10) to (12) or by virtue of subsections (13) and (14), if the court considers that setting it aside would be contrary to the interests of justice.
    (16) In this section, the expression "offence against the course of justice"—
    (a) means an offence of perverting, or of attempting to pervert, the course of justice (by whatever means and however the offence is described), and
    (b) without prejudice to the generality of paragraph (a), includes—
    (i) an offence under section 45(1) of the Criminal Law (Consolidation) (Scotland) Act 1995 (c.39) (aiding, abetting, counselling, procuring or suborning the commission of an offence under section 44 of that Act),
    (ii) the crime of subornation of perjury, and
    (iii) the crime of bribery.
    (17) But the expression does not include—
    (a) the crime of perjury, or
    (b) an offence under section 44(1) of that Act (statement on oath which is false or which the person making it does not believe to be true).
    NOTE
    Section 4 implements Recommendations 13, 14, 16 to 20 and 22 to 23 on tainted acquittals (see paras 3.1-3.52 and 3.55-3.61). A person who has been acquitted of an offence – whether on indictment or in summary proceedings – may be tried again for that offence, or for another offence arising out of the same, or largely the same, acts and omissions, if the conditions set out in subsection (3) are satisfied. The first condition is that either some person has been convicted of an offence against the course of justice in connection with the original proceedings, or that the High Court, on an application by the Lord Advocate, has concluded on the balance of probabilities that such an offence was committed. The second condition is that on the application of the Lord Advocate the High Court has set aside the acquittals and granted authority to bring a new prosecution.
    Any application under subsection (3)(a)(ii) or (3)(b) must be heard by a court of three judges (subsection 4), whose decision on the application is final (subsection 7). A copy of the application must be sent to the acquitted person (subsection (4)) who is entitled to appear or be represented at any hearing on the application (subsection (5)). The court may appoint counsel to act as amicus curiae at the hearing; this would be particularly important in any case in which the offence against the course of justice was allegedly committed by a third party, where there might otherwise be no-one in a position to contradict the Lord Advocate's application (see para 3.52).
    The section provides for two tests. If the offence against the course of justice involved interference with a juror or with the trial judge, the court is to set aside the acquittal if it is unable to conclude that the interference had no effect on the outcome of the original proceedings (subsections (8) and (9)). An acquittal is not to be set aside, however, where interference with the jury became known to the trial judge, who then allowed the trial to proceed to its conclusion. Where the offence against the course of justice did not involve interference with a juror or with the trial judge, the court may set aside the acquittal only if it is satisfied on the balance of probabilities that the interference led to the withholding of evidence, or to the giving of false evidence, which was likely to have had a material effect on the outcome of the original proceedings. This test is discussed at paras 3.36-3.48.
    For the purposes of the section, an "offence against the course of justice" is defined as meaning an offence of perverting, or attempting to pervert, the course of justice (subsection (16)), but as excluding the crime of perjury and its statutory analogue, an offence under section 44(1) of the 1995 Act. See the discussion at paras 3.9-3.14.
    5 Further provision as regards prosecution by virtue of section 4
    (1) A prosecution may be brought by virtue of section 4 notwithstanding that any time limit for the commencement of such proceedings has elapsed.
    (2) In proceedings in a prosecution brought by virtue of section 4 it is competent for either party to lead evidence which it was competent for that party to lead in the earlier proceedings.
    (3) But the indictment or complaint in the prosecution is to identify any matters as respects which the prosecutor intends to lead evidence by virtue of subsection (2) (being matters as respects which it would not have been competent to lead evidence but for that subsection).
    (4) On granting authority under section 4(3)(b)(ii) to bring a new prosecution, the High Court may, after giving the parties an opportunity of being heard, order the detention of the accused person in custody or admit that person to bail.
    (5) In—
    (a) solemn proceedings, section 65(4)(aa) and (b) and (4A) to (9), and
    (b) summary proceedings, section 147,
    of the 1995 Act (prevention of delay in trials) applies to an accused person who is detained under subsection (4) as it applies to an accused person detained by virtue of being committed until liberated in due course of law.
    NOTE
    Section 5 contains a number of technical provisions governing procedure in a prosecution authorised by virtue of section 4. In common with sections 7 and 9, section 5 is modelled upon section 119 (provision where the High Court authorises new prosecution) of the 1995 Act. In contrast to section 119, section 5 does not limit the offence which may be charged, or the penalty which may be applied, by reference to the charge at the original trial (cf s 119(2), (3)).
    6 Admission subsequent to acquittal
    (1) A person who, whether on indictment or complaint (the "original" indictment or complaint), has been acquitted of an offence but subsequently admits to committing it may, provided that the condition mentioned in subsection (3) is satisfied, be charged with, and prosecuted anew for, the offence.
    (2) Whether the acquittal was before or after the coming into force of this section is, for the purposes of the section, immaterial.
    (3) The condition is that on the application of the Lord Advocate the High Court has—
    (a) set aside the acquittal, and
    (b) granted authority to bring, by virtue of this section, a new prosecution.
    (4) On making an application under subsection (3), the Lord Advocate is to send a copy of that application to the acquitted person.
    (5) The acquitted person is entitled to appear or to be represented at any hearing of the application.
    (6) For the purpose of hearing and determining the application, three of the Lords Commissioners of Justiciary are a quorum of the Court (the application being determined by majority vote of those sitting).
    (7) An acquittal is not to be set aside under subsection (3)(a) unless the Court is satisfied—
    (a) on a balance of probabilities, that subsequent to the acquittal the person credibly admitted having committed the offence, and
    (b) that evidence is available sufficient to corroborate the admission.
    (8) Even if the Court is satisfied as is mentioned in subsection (7), it is not to set aside the acquittal if it considers that to do so would be contrary to the interests of justice.
    NOTE
    Section 6 provides for the retrial of a person who is acquitted of an offence and then admits to having committed it (Recommendation 25, paras 4.2-4.7). It applies regardless of whether the original acquittal was obtained in solemn or summary proceedings (subsection (1)) and regardless of whether the original acquittal was obtained prior to the coming into force of the section (subsection (2), implementing Recommendation 28, para 4.8-4.11).
    An application for the setting aside of the acquittal and for authority to bring a new prosecution may be made by the Lord Advocate, the application being determined by a court of three judges (subsections (3) and (4), implementing Recommendation 26, para 4.7).
    The court may set aside the acquittal only if satisfied on a balance of probabilities that the person credibly admitted having committed the offence and that evidence is available which is sufficient to corroborate the admission (subsection (5)), and may not do so where it considers that setting aside the acquittal would be contrary to the interests of justice (subsection (6),implementing Recommendation 27).
    7 Further provision as regard prosecution by virtue of section 6
    (1) No sentence may be passed on conviction in a new prosecution brought by virtue of section 6 which could not have been passed under the proceedings on the original indictment or complaint ("the earlier proceedings").
    (2) A new prosecution may be brought by virtue of section 6 notwithstanding that any time limit, other than the time limit mentioned in subsection (3), for the commencement of such proceedings has elapsed.
    (3) Proceedings in a new prosecution brought by virtue of section 6 are to be commenced within 2 months after the date on which authority to bring the prosecution was granted.
    (4) In proceedings in a new prosecution brought by virtue of section 6 it is competent for either party to lead evidence which it was competent for that party to lead in the earlier proceedings.
    (5) But the indictment or complaint in the new prosecution is to identify any matters as respects which the prosecutor intends to lead evidence by virtue of subsection (4) (being matters as respects which it would not have been competent to lead evidence but for that subsection).
    (6) For the purposes of subsection (3), proceedings are deemed commenced—
    (a) in a case where a warrant to apprehend the accused is granted—
    (i) on the date on which the warrant is executed, or
    (ii) if it is executed without unreasonable delay, on the date on which it is granted, and
    (b) in any other case, on the date on which the accused is cited.
    (7) Where the 2 months mentioned in subsection (3) elapse and no new prosecution has been brought under this section, the order under section 6(3)(a) setting aside the acquittal has the effect, for all purposes, of an acquittal.
    (8) On granting authority under section 6(3)(b) to bring a new prosecution, the High Court may, after giving the parties an opportunity of being heard, order the detention of the accused person in custody or admit that person to bail.
    (9) In—
    (a) solemn proceedings, section 65(4)(aa) and (b) and (4A) to (9), and
    (b) summary proceedings, section 147,
    of the 1995 Act (prevention of delay in trials) applies to an accused person who is detained under subsection (8) as it applies to an accused person detained by virtue of being committed until liberated in due course of law.
    (10) It is immaterial, for the purposes of this section, whether the acquittal was before or after the coming into force of the section.
    NOTE
    Section 7 contains a number of technical provisions governing procedure in a prosecution authorised by virtue of section 6. In common with sections 5 and 9, section 7 is modelled upon section 119 (provision where the High Court authorises new prosecution) of the 1995 Act.
    8 New evidence
    (1) A person who has been acquitted, after the coming into force of this section (or on the day on which it comes into force), of an offence may—
    (a) if there is new evidence that the person committed the offence, and
    (b) the conditions mentioned in subsection (2) are satisfied,
    be charged with, and prosecuted for, the offence anew.
    (2) The conditions are—
    (a) that the person's acquittal was of an offence mentioned in subsection (9), and
    (b) that on the application of the Lord Advocate the High Court has—
    (i) set aside the acquittal, and
    (ii) granted authority to bring, by virtue of this section, a new prosecution.
    (3) The setting aside of the acquittal and the granting of such authority may, under subsection (2)(b), be applied for on one occasion only.
    (4) On making an application under that subsection, the Lord Advocate is to send a copy of the application to the acquitted person.
    (5) The acquitted person is entitled to appear or to be represented at any hearing of the application.
    (6) For the purpose of hearing and determining the application under subsection, three of the Lords Commissioners of Justiciary are a quorum of the Court (the application being determined by majority vote of those sitting).
    (7) An acquittal is not to be set aside under subsection (2)(b)(i) unless the Court is satisfied that—
    (a) the case against the accused is strengthened substantially by the new evidence,
    (b) the new evidence is evidence which was not available, and could not with the exercise of reasonable diligence have been made available, at the trial in respect of the original offence, and
    (c) on the new evidence and the evidence which was led at that trial it is highly likely that a reasonable jury properly instructed would have convicted the person of the offence.
    (8) Even if the Court is satisfied as is mentioned in subsection (7), it is not to set aside the acquittal if it considers that to do so would be contrary to the interests of justice.
    (9) The offences are—
    (a) murder,
    (b) at common law, rape, and
    (c) an offence under either section 1 (rape) or section 18 (rape of a young child) of the Sexual Offences (Scotland) Act 2009 (asp 9).
    (10) The Scottish Ministers may, by order made by statutory instrument, amend subsection (9) so as to add further offences to those for the time being mentioned in that subsection.
    (11) But subsection (1) does not apply as respects a person's acquittal of an offence so added if the date of acquittal is earlier than that on which the addition is effected.
    (12) An order under subsection (10) is not made unless a draft of the statutory instrument containing the order has been—
    (a) laid before, and
    (b) approved by a resolution of,
    the Parliament.
    NOTE
    Section 8 provides for an exception to the rules in sections 1 and 2 on the basis of new evidence.
    The exception applies only in relation to the offences specified in subsection (9). The offences are murder and rape (whether at common law or under section 1 or 18 of the Sexual Offences (Scotland) Act 2009) (Recommendation 29, paras 5.2 – 5.6). Subsection (10) gives the Scottish Ministers power, by way of an order made by statutory instrument subject to affirmative resolution of the Scottish Parliament, to amend subsection (9) so as to add further offences (Recommendation 30, paras 5.2 – 5.6). Any such amendment will apply the section only in relation to acquittals delivered after the amendment's coming into force.
    Application for the setting aside of the acquittal and for authority to bring a new prosecution may be made by the Lord Advocate to a court consisting of three judges (subsections (2)(b), (3)). The application may be granted only if satisfied that the case against the accused is strengthened substantially by the new evidence, that the new evidence is evidence which was not available, and could not with the exercise of reasonable diligence have been made available, at the trial in respect of the original offence, and that on the new evidence and the evidence which was led at the original trial it is highly likely that a reasonable jury properly instructed would have convicted the person of the offence. (This test implements Recommendations 31-33; for discussion, see paras 5.14-5.20 and 5.56-5.62). The court may not grant the application where it considers that to do so would be contrary to the interests of justice (subsection 8, implementing Recommendation 34).
    9 Further provision as regards prosecution by virtue of section 8
    (1) No sentence may be passed on conviction in a new prosecution brought by virtue of section 8 which could not have been passed under the indictment on the trial of which the person was acquitted of the offence in question.
    (2) A new prosecution may be brought by virtue of section 8 notwithstanding that any time limit for the commencement of such proceedings, other than the time limit mentioned in subsection (3), has elapsed.
    (3) Proceedings in a new prosecution brought by virtue of section 8 are to be commenced within 2 months after the date on which authority to bring the prosecution was granted.
    (4) In proceedings in a new prosecution brought by virtue of section 8 it is competent for either party to lead evidence which it was competent for that party to lead in the earlier proceedings.
    (5) But the indictment in the new prosecution is to identify any matters as respects which the prosecutor intends to lead evidence by virtue of subsection (4) (being matters as respects which it would not have been competent to lead evidence but for that subsection).
    (6) For the purposes of subsection (3), proceedings are deemed commenced—
    (a) in a case where a warrant to apprehend the accused is granted—
    (i) on the date on which the warrant is executed, or
    (ii) if it is executed without unreasonable delay, on the date on which it is granted, and
    (b) in any other case, on the date on which the accused is cited.
    (7) Where the 2 months mentioned in subsection (3) elapse and no new prosecution has been brought under this section, the order under section 8(2)(b)(i) setting aside the acquittal has the effect, for all purposes, of an acquittal.
    (8) On granting authority under section 8(2)(b)(ii) to bring a new prosecution, the High Court is, after giving the parties an opportunity of being heard, to order the detention of the accused person in custody or to admit that person to bail.
    (9) Subsections (4)(aa) and (b) and (4A) to (9) of section 65 of the 1995 Act (prevention of delay in trials) apply to an accused person who is detained under subsection (8) as they apply to an accused person detained by virtue of being committed until liberated in due course of law.
    NOTE
    Section 9 contains a number of technical provisions governing procedure in a prosecution authorised by virtue of section 8. In common with sections 5 and 7, section 9 is modelled upon section 119 (provision where the High Court authorises new prosecution) of the 1995 Act.
    10 Nullity of proceedings on previous indictment or complaint
    (1) Subsection (3) applies where—
    (a) a person has, whether on indictment or complaint—
    (i) been charged with, and
    (ii) acquitted or convicted of,
    an offence, and
    (b) the conditions mentioned in subsection (4) are satisfied.
    (2) Whether the conviction or acquittal was before or after the coming into force of this section is, for the purposes of the section, immaterial.
    (3) The person may be charged with, and prosecuted anew for, the offence.
    (4) The conditions are that, on the application of the prosecutor and after hearing the parties, the High Court is satisfied—
    (a) that the proceedings on the indictment or complaint were a nullity, and
    (b) that it would not be contrary to the interests of justice to proceed as mentioned in subsection (3).
    NOTE
    Section 10 requires the Lord Advocate, in any case where the Crown would propose to argue in response to any issue raised under section 1 or 2 that the original proceedings were a nullity, to apply to the High Court before bringing a prosecution. This is aimed at the very small number of cases in which the first proceedings are a fundamental nullity and so cannot be regarded as leading to a valid acquittal or conviction (see paras 2.76-2.84).
    11 Amendment of Schedule 1 to the Contempt of Court Act 1981
    (1) Schedule 1 to the Contempt of Court Act 1981 (c.49) (times when proceedings are active for the purposes of section 2 of that Act) is amended as follows.
    (2) After paragraph 1 (the expressions "criminal proceedings" and "appellate proceedings"), there is inserted—
    "1A Proceedings under the Double Jeopardy (Scotland) Act 2009 (asp00) are criminal proceedings (and are not appellate proceedings) for the purposes of this Schedule.".
    (3) In paragraph 4 (initial steps of criminal proceedings), at the end there is added—
    "(f) the making of an application under section 4(3)(a)(ii) or (b) (tainted acquittals), 6(3) (admission subsequent to acquittal) or 8(2)(b) (new evidence) of the Double Jeopardy (Scotland) Act 2009 (asp00).".
    (4) In paragraph 5 (conclusion of criminal proceedings), at the end there is added—
    "(d) where the initial steps of the proceedings are as mentioned in paragraph 4(f)—
    (i) by refusal of the application,
    (ii) if the application is granted and within 2 months thereafter a new prosecution is brought, by acquittal, or as the case may be by sentence, in the new prosecution.".
    (5) In paragraph 7 (discontinuance of proceedings), at the end there is added—
    "(d) where the initial steps of the proceedings are as mentioned in paragraph 4(f) and the application is granted, if no new prosecution is brought within 2 months thereafter.".
    NOTE
    Section 11 amends Schedule 1 to the Contempt of Court Act 1981 to bring an application under section 4, 6 or 8 of the Bill and any retrial following thereon within the protection provided by that act to active criminal proceedings. This implements Recommendations 22 and 35. The effect of the amendment is that criminal proceedings will be taken to be active from the making of an application under section 4, 6 or 8 until the refusal of that application or, if the application is granted and a new trial is commenced within two months, until the termination of that trial in acquittal or sentence. During this period, the rule of strict liability contempt in section 2 of the 1981 Act will apply in relation to any publication which creates a substantial risk of prejudicing the proceedings. In addition, the court will have have power to make an order under section 4(2) of that Act postponing publication of reports of the proceedings.
    12 Short title, interpretation and commencement
    (1) This Act may be cited as the Double Jeopardy (Scotland) Act 2009.
    (2) In this Act, "the 1995 Act" means the Criminal Procedure (Scotland) Act 1995 (c.46).
    (3) This Act, except this section, comes into force on such day as the Scottish Ministers may by order made by statutory instrument appoint.
    Appendix B List of Respondents
    Association of Chief Police Officers in Scotland
    Crown Office and Procurator Fiscal Service ("COPFS")
    Faculty of Advocates
    Gerard Sinclair
    James Chalmers, University of Edinburgh
    Keir Starmer QC, Director of Public Prosecutions
    The Criminal Committee of the Law Society of Scotland
    Michael Christie
    Professor Peter Duff and Dr Liz Campbell, University of Aberdeen
    PW Ferguson QC
    Professor Paul Roberts
    The Judges of the High Court of Justiciary
    Robin Bennett, Bennetts Solicitors
    Uganda Law Reform Commission

Note 1   Scot Law Com No 212. This was preceded by a Discussion Paper (DP 137), published in March 2008.    [Back]

Note 2   We are grateful to all those who responded to the Discussion Paper. Appendix B contains a list of respondents.    [Back]

Note 3   Scotland Act 1998, s 126(5); Sch 5.    [Back]

Note 4   Ibid, s 29(2)(d). For the definition of "Convention rights" see ibid, s 126(1) and, by reference, s 1 of the Human Rights Act 1998.    [Back]

Note 5   That is, Art 54 of the Convention Implementing the Schengen Agreement of 19 June 1990 (OJ L 239, 22.10.2000, pp 19-62).    [Back]

Note 6   For a discussion of the Schengen Convention, as interpreted by the European Court of Justice, see Part 2 of Appendix 1 to the Discussion Paper.    [Back]

Note 7   Nikitin v Russia (2005) 41 EHRR 10 at para 57.    [Back]

Note 8   These matters are considered in depth in Part 1 of Appendix 1 to the Discussion Paper.    [Back]

Note 9   See paras 3.59-3.63 below.    [Back]

Note 10   Discussion Paper, paras 2.30-2.33.    [Back]

Note 11   Ibid, paras 2.32-2.37.    [Back]

Note 12   Hume, ii, 465; a fuller quotation may be found at para 3.5 of the Discussion Paper.    [Back]

Note 13   For a summary of the existing cases, see paras 3.31-3.37 of the Discussion Paper.    [Back]

Note 14   See paras 3.21-3.28 of the Discussion Paper.    [Back]

Note 15   Cf the civil principle of res judicata as explained by Lord President Cooper in Grahame v Secretary of State for Scotland 1951 SC 368 at 387.    [Back]

Note 16   More precisely, it is never appropriate to try someone again on a charge in respect of which there is a valid subsisting verdict: so a second prosecution may be appropriate either where the first proceedings were a nullity (Discussion Paper, para 3.29) or where the original conviction has been quashed on appeal and authority for a new prosecution granted in terms of ss 118(1)(c) and 119 of the Criminal Procedure (Scotland) Act 1995 or their equivalents in summary procedure.    [Back]

Note 17   Discussion Paper, para 6.35.    [Back]

Note 18   Discussion Paper, para 6.37, proposal 3.    [Back]

Note 19   That is (in appeals against conviction in solemn proceedings) ss 118(1)(c) and 119 and (in appeals by stated case) ss 183(1)(d) and 185.    [Back]

Note 20   Discussion Paper, para 6.6.    [Back]

Note 21   Ex hypothesi, the actual facts have not been proved, so the court would have to base its assessment of whether or not the facts were the same upon the averred facts. The test thus becomes very sensitive to the way in which the first indictment was drawn.    [Back]

Note 22   (1865) 5 Irv 203.    [Back]

Note 23   (1863) 4 Irv 444; Discussion Paper para 3.34.    [Back]

Note 24   Discussion Paper, Appendix 1, Part 2.    [Back]

Note 25   Van Esbroeck, Case C-436/04, [2006] ECR I-2333 at para 36.    [Back]

Note 26   Ibid, at para 38.    [Back]

Note 27   1980 JC 111.    [Back]

Note 28   A note on terminology: both the Faculty of Advocates and COPFS pointed out that one would normally speak not of separation of trials but of separation of charges. What we meant by the term "separation of trials" was the circumstance in which the accused sought to have one or more of the charges against him tried in separate proceedings from the others.    [Back]

Note 29   Only the Faculty of Advocates explicitly endorsed the first option of a rule with limited exceptions.    [Back]

Note 30   1984 SLT 345.    [Back]

Note 31   See para 2 of Sch 3 to the 1995 Act.    [Back]

Note 32   Cf R v IK [2007] EWCA Crim 971, [2007] 2 Cr App R 15, in which it was held that the prior conviction of the appellants for forgery offences did not render it an abuse of process subsequently to prosecute them for terrorist offences of which the forgery formed a part.     [Back]

Note 33   Isabella Cobb or Fairweather (1836) 1 Swin 354.    [Back]

Note 34   1994 SCCR 451 at 453.    [Back]

Note 35   See Recommendation 3 at para 2.18 above.     [Back]

Note 36   At para 6.51.    [Back]

Note 37   Expressed at para 6.57 of the Discussion Paper.    [Back]

Note 38   As we observed at para 6.51 of the Discussion Paper.    [Back]

Note 39   As set out in para 2.38 above.    [Back]

Note 40   This point is even clearer in the context of certain statutory offences: even a cursory reading of the Road Traffic Act 1988 shows that causing death by dangerous driving (s 1) is a separate offence, and not merely an aggravated form of dangerous driving (s 2).    [Back]

Note 41   That is, those that do not involve an assault upon the deceased.    [Back]

Note 42   Section 3 of the Road Traffic Act 1998. Following the death of the victim, one might in appropriate circumstances wish to prosecute the driver for causing death by careless driving while under the influence of drink or drugs (s 3A of that Act).     [Back]

Note 43   Para 6.53, fn 59.    [Back]

Note 44   Principally, but not exclusively. Death may also result from common law offences, such as fire-raising, which do not involve an assault. At the time of writing, the question of whether evil intent is a required part of the mens rea of murder or whether wicked recklessness will suffice is uncertain; the question is expected to be considered shortly by a bench of five judges in Petto v HMA (last reported as [2009] HCJAC 43; 2009 SLT 509).    [Back]

Note 45   Part 1 of Sch 2 to the Road Traffic Offenders Act 1988.    [Back]

Note 46   See the proposal to this effect in PW Ferguson, "Double Jeopardy" 2009 SCL 669 at 677.    [Back]

Note 47   Discussion Paper, paras 3.24-3.25.    [Back]

Note 48   (1903) 4 Adam 124.    [Back]

Note 49   Discussion Paper, para 6.64.    [Back]

Note 50   Discussion Paper, paras 3.44-3.45.    [Back]

Note 51   Discussion Paper, Appendix 1, Part 2.    [Back]

Note 52   Discussion Paper, paras 6.70-6.71.    [Back]

Note 53   Yeung Chun Pong and Others v Secretary for Justice [2009] HKCFA 57, at para 21.    [Back]

Note 54   As it does in the United States of America: Bartkus v Illinois 359 US 121 (1959).    [Back]

Note 55   1974 SLT (Notes) 42.    [Back]

Note 56   See too HMA v Joseph 1929 JC 55, in which it was held competent in an indictment for a number of frauds committed in Scotland to narrate, and then to lead evidence in relation to, a similar offence allegedly committed in Belgium.    [Back]

Note 57   Discussion Paper, paras 3.29-3.30. There we said that, in order to found a plea ofres judicata, the original proceedings must have beencompetent. This was too broad a statement, since the verdict of incompetent proceedings, if not set aside, may nevertheless ground a plea of tholed assize: Robertson v Higson 2006 SC (PC) 22, per Lord Rodger of Earlsferry at para 43.    [Back]

Note 58   See, for instance, Island Maritime Ltd v Filipowski (2006) 228 ALR 1 (High Court of Australia).    [Back]

Note 59   1974 JC 50.    [Back]

Note 60   1993 SLT 599.    [Back]

Note 61   That section provided that where a person had been charged on indictment and the trial had not been commenced within 12 months of that person's first appearance on petition, "the accused shall be discharged forthwith and thereafter he shall be for ever free from all question or process for that offence". In Whitelaw, the High Court held that this provision barred only further proceedings on indictment, but did not bar further proceedings on complaint. This was overruled by a Full Bench in Gardner v Lees 1996 JC 83.    [Back]

Note 62   1996 SCCR 879.    [Back]

Note 63   1996 JC 224.    [Back]

Note 64   (1824), Alison at 619.    [Back]

Note 65   Cf the opinion of a Full Bench, and later the Judicial Committee of the Privy Council, in relation to the effect of verdicts of temporary sheriffs which, although admittedly incompetent, had not been challenged timeously: Robertson v Higson 2005 HCJAC 2, 2005 JC 210; 2006 SC (PC) 22.     [Back]

Note 66   Stuurman v HMA 1980 JC 111.    [Back]

Note 67   Discussion Paper, paras 7.5-7.14.    [Back]

Note 68   138 F 3d 302 (7th cir, 1998).    [Back]

Note 69   Ibid, at para 19.    [Back]

Note 70   Ibid, at paras 22-23.    [Back]

Note 71   Discussion Paper, para 7.15, question 14.    [Back]

Note 72   Ibid, paras 7.20-7.21, question 22.    [Back]

Note 73   What constitutes such a taint is considered at paras 3.9-3.12, 3.36-3.40 and 3.42-3.46 below.    [Back]

Note 74   Cf the right to a fair trial under Art 6(1) ECHR.    [Back]

Note 75   Discussion Paper, para 7.24, referring to HMA v Cairns 1967 JC 37.    [Back]

Note 76   Ibid, para 7.25, proposal 24.    [Back]

Note 77   1967 JC 37.    [Back]

Note 78   See below, para 4.2-4.6.    [Back]

Note 79   Discussion Paper, para 7.15, question 15.    [Back]

Note 80   Ibid, question 16.    [Back]

Note 81   Ibid, question 17.    [Back]

Note 82   Ibid, question 18.    [Back]

Note 83   Ibid, question 19.    [Back]

Note 84   Discussion Paper, paras 3.29-3.30. This seems to have been the option favoured by COPFS, who suggested that it might be appropriate simply for the (second) trial court to have discretion to prevent the trial on grounds of oppression following the service of a new indictment.    [Back]

Note 85   Hume, ii, 468, referring to the cases of John Wallace (1730) and Janet Macrachan (1758). A similar approach was hinted at by Lord Hoffman in giving the opinion of the Judicial Committee of the Privy Council in Forbes v Attorney General of Jamaica [2009] UKPC 13 at para 13.    [Back]

Note 86   Criminal Procedure and Investigations Act 1996, s 54(1)(b).    [Back]

Note 87   Ibid, s 54(2)(b).    [Back]

Note 88   Ibid, s 55(1).    [Back]

Note 89   Cf Thomson v Crowe 2000 JC 173 at 192B-F. Cf also Aleman, 138 F 3d 302 (7th cir, 1998) (discussed at para 3.2 above) in which the US Court of Appeals for the 7th Circuit made no criticism of the statement of the court below that the State's burden was to establish the bribery by a preponderance of the evidence.     [Back]

Note 90   Although this risk may be managed by reporting restrictions: see para 3.55 below.    [Back]

Note 91   2002 JC 99 at para 219 (Opinion of the Court delivered by Lord Justice General Cullen).    [Back]

Note 92   Where the trial features a single accused it is safe to assume that there will be no intimidation or bribery aimed at preventing a defence witness from giving evidence: if the accused did not wish the evidence of a defence witness to be heard he would simply refrain from citing that witness. However, the possibility of the intimidation or bribery of defence witnesses must arise in cases where co-accused run "cut-throat" defences of incrimination.    [Back]

Note 93   See para 3.34 above.    [Back]

Note 94   Cf the 18th century cases of corruption among justices and magistrates cited by Hume and referred to at para 3.30 of the Discussion Paper.    [Back]

Note 95   Cf Stewart v HMA 1980 JC 103, an appeal against conviction in which the trial judge had dismissed a juror after learning that she had been offered a bribe by a person claiming to be the brother of the accused.    [Back]

Note 96   PW Ferguson, "Double Jeopardy" 2009 SCL 669 at 674.    [Back]

Note 97   Discussion Paper, para 7.15, question 18.    [Back]

Note 98   Para 7.21, questions 20, 21.    [Back]

Note 99   Hume, ii, 468-9: "[C]ertainly the absolivitor can only be available to the pannels, so far as it has been obtained in a fair and regular process, truly intended to answer the purposes of justice, and not to screen him from punishment, under colour of a previous trial. Thus, John Wallace was indicted at instance of the Lord Advocate, for assaulting and abusing an officer of the revenue. But, said he, I have already been tried and acquitted, on a complaint at instance of the procurator-fiscal to the Justices of the Peace for the shire of Forfar, where the thing was done. In answer, the prosecutor affirmedinter alia, that this had been a mere collusive trial, truly intended to protect the prisoner from the due consequences of his crime. Accordingly, the Lords paid no regard to this pretendedres judicata."    [Back]

Note 100   Law Com No 267 (2001).    [Back]

Note 101   Para 7.27.    [Back]

Note 102   1967 JC 37, at 46-47.    [Back]

Note 103   [2002] HCA 57.    [Back]

Note 104   As shown in Miell [2007] EWCA Crim 3130; [2008] 1 WLR 627.    [Back]

Note 105   Ibid.    [Back]

Note 106   [2006] EWCA Crim 1354; [2007] 1 WLR 1657.    [Back]

Note 107   [2009] EWCA Crim 633.    [Back]

Note 108   Article 7(1) ECHR 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 . . . at the time it was committed. Nor shall a heavier penalty be imposed than the one that was applicable at the time the criminal offence was committed."    [Back]

Note 109   Discussion Paper, para 7.33.    [Back]

Note 110   Ibid, para 7.39.    [Back]

Note 111   See the Home Affairs Committee, Third Report of Session 1999-2000, The Double Jeopardy Rule, HC Paper No 190 (Session 1999/2000) at para 20.    [Back]

Note 112   The applications are R v Dunlop [2006] EWCA Crim 1354, [2007] 1 WLR 1657; R v Miell [2007] EWCA Crim 3130, [2008] 1 WLR 627; R v Andrews [2008] EWCA Crim 2908, [2009] 1 WLR 1947; R v B(J) [2009] EWCA Crim 1036, [2009] Crim LR 736; R v G(G) [2009] EWCA Crim 1207, [2009] Crim LR 738; and R v Celaire [2009] EWCA Crim 633 (unreported, 17 March 2009). The applications in Miell, B(J) and G(G) were refused.    [Back]

Note 113   Mr Gerard Sinclair.    [Back]

Note 114   In particular, the statistics referred to in para 4.16 above.    [Back]

Note 115   Home Affairs Committee, Third Report of Session 1999-2000, The Double Jeopardy Rule, HC Paper No 190 (Session 1999/2000) at para 27, quoted in David Hamer, "The expectation of incorrect acquittals and the 'new and compelling evidence' exception to double jeopardy" (2009) Crim LR 63. Regarding ear prints, see R v Kempster [2008] EWCA Crim 975, [2008] 2 Cr App R 19.    [Back]

Note 116   Professor Roberts, in his response to the Discussion Paper, said that "[r]eferences to corneal mapping and the like are mostly science fiction for present purposes."     [Back]

Note 117   Hume, ii, 465.    [Back]

Note 118   Green v United States 355 US 184 (1957) at 187-188.    [Back]

Note 119   Discussion Paper, para 7.43. The Law Commission recommended that the exception should only apply in cases of murder and genocide consisting in the killing of any person: Law Commission, Double Jeopardy and Prosecution Appeals, Law Com No 267 (2001), para 4.42.    [Back]

Note 120   Criminal Justice Act 2003, Sch 5, Pt 1.    [Back]

Note 121   Discussion Paper, para 7.43, question 27.    [Back]

Note 122   Criminal Procedure (Scotland) Act 1995, s 205.    [Back]

Note 123   The reasons for this restriction are substantially the same as those that apply in relation to retrospection more generally: see paras 5.71-5.84 below.    [Back]

Note 124   Criminal Justice Act 2003, s 78(2).    [Back]

Note 125   Ibid, s 79(2)(c).    [Back]

Note 126   At paras 7.48-7.49.    [Back]

Note 127   Application 65582/01 (judgment 24 May 2007), at para 50.     [Back]

Note 128   An example of this may be seen in R v Andrews [2008] EWCA Crim 2908; [2009] 1 Cr App R 26, in which the new evidence appears to have become admissible only following changes to the rules on similar fact evidence introduced by the Criminal Justice Act 2003: see para 33 of Andrews.    [Back]

Note 129   Gerard Sinclair.    [Back]

Note 130   Section 106(3A).    [Back]

Note 131   For a recent example, see Fraser v HMA [2008] HCJAC 26, 2008 SCCR 407 at paras 139-144.    [Back]

Note 132   Section 106(3B).    [Back]

Note 133   Law Com CP No 156 (1999), paras 5.47-5.48.    [Back]

Note 134   Discussion Paper, para 7.54.    [Back]

Note 135   Law Com CP No 156 (1999), paras 5.34-5.42.    [Back]

Note 136   Law Commission, Double Jeopardy and Prosecution Appeals, Law Com No 267 (2001) at para 4.62 (emphasis in original).    [Back]

Note 137   Ibid at para 4.65 (emphasis in original).    [Back]

Note 138   [2006] EWCA Crim 1354; [2007] 1 WLR 1657.    [Back]

Note 139   [2007] EWCA Crim 3130; [2008] 1 WLR 627.    [Back]

Note 140   [2006] EWCA Crim 1354; [2007] 1 WLR 1657.    [Back]

Note 141   [2007] EWCA Crim 3130; [2008] 1 WLR 627.    [Back]

Note 142   Ibid, at para 38.    [Back]

Note 143   Ibid, at paras 42-45.    [Back]

Note 144   [2008] EWCA Crim 2908; [2009] 1 Cr App R 26.    [Back]

Note 145   [2008] EWCA Crim 2908; [2009] 1 Cr App R 26, at para 36.    [Back]

Note 146   Ibid, at para 38.    [Back]

Note 147   [2009] EWCA Crim 1036, [2009] Crim LR 736.    [Back]

Note 148   Section 74 provides for convicted persons to secure sentence discounts for providing assistance to the prosecutor or investigator of an offence.     [Back]

Note 149   [2009] EWCA Crim 1036, para 9.    [Back]

Note 150   [2009] EWCA Crim 1207, [2009] Crim LR 738.    [Back]

Note 151   [2009] EWCA Crim 633.    [Back]

Note 152   Law Commission, Double Jeopardy and Prosecution Appeals, Law Com No 267 (2001), para 4.61. (Emphasis in original.)    [Back]

Note 153   See para 5.51 below.    [Back]

Note 154   2002 JC 99, para 219. See above, paras 3.34-3.35.    [Back]

Note 155   1987 SCCR 608; 1988 SLT 169.    [Back]

Note 156   Cf the Law Commission's discussion of an analogous suggestion in England and Wales: "The test recommended by the Home Affairs Select Committed was whether the new evidence makes the acquittal 'unsafe'. . . 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." Law Commission, Double Jeopardy and Prosecution Appeals Law Com No 267 (2001), para 4.66. Although the Scottish test of miscarriage of justice is more exacting than the English test of safety of the conviction, excluding as it does the possibility of a successful appeal on the basis of 'lurking doubt', the Law Commission's argument appliesmutatis mutandis.    [Back]

Note 157   We are here concerned only with the test of miscarriage of justice in appeals under solemn procedure. The nature of the appeal court's role in an appeal by stated case in summary procedure is quite distinct, by virtue of the detail which the judge in summary proceedings is able to provide of the view which he took of the evidence. In such cases, where the reasoning of the fact-finder is not inscrutable, it is appropriate to refer to the prosecutor's right of appeal as being in respect of a "miscarriage of justice": cf s 175(5E) of the 1995 Act.    [Back]

Note 158   Fraser v HMA [2008] HCJAC 26, 2008 SCCR 407 at para 211.    [Back]

Note 159   Cf the opinion of the Court of Appeal in R v Andrews [2008] EWCA Crim 2908; [2009] 1 Cr App R 26 at [41]: ". . . the principle of finality in litigation does not, as a principle, provide a relevant consideration bearing on the interests of justice. Double jeopardy as a prohibition against a second trial following an acquittal was abolished by the [Criminal Justice Act 2003]."    [Back]

Note 160   Such a balancing exercise seems to have been contemplated by the High Court in X v Sweeney 1982 JC 70: see, in particular, Lord Justice General Emslie at 85, Lord Cameron at 87 and Lord Avonside at 92.    [Back]

Note 161   Montgomery v HMA 2001 SC (PC) 1, per Lord Hope of Craighead at 29.    [Back]

Note 162   1980 JC 111.    [Back]

Note 163   See, for instance, Atkins v London Weekend Television Ltd 1978 JC 48, Spink v HM Advocate 1989 SCCR 413, McLeod v HM Advocate 1998 SLT 60.    [Back]

Note 164   For instance in R v McCann (1991) 92 Cr App R 239; R v Taylors (1994) 98 Cr App R 361. In Atkins, while a later prosecution was technically competent, no new indictment was in fact served.    [Back]

Note 165   Cf R v Dunlop [2006] EWCA Crim 1354, [2007] 1 Cr App R 8, in which the Court of Appeal noted (at para 26) that "any recollection that members of the jury might have in relation to publicity about Dunlop would pale into insignificance in comparison to the legitimate prejudicial effect of being told that he had, on a number of occasions, confessed to her murder and that he had pleaded guilty to perjury in relation to his denial of being guilty of that offence."    [Back]

Note 166   Cf Criminal Procedure (Scotland) Act 1995, s 119(2), (3).    [Back]

Note 167   Law Com No 267 (2001) at para 4.53.    [Back]

Note 168   Article 8 provides: "(1) Everyone has the right to respect for his private and family life, his home and his correspondence. (2) There shall be no interference by a public authority with the exercise of this right except such as is in accordance with the law and is necessary in a democratic society in the interests of national security, public safety or the economic wellbeing of the country, for the prevention of disorder or crime, for the protection of health or morals, or for the protection of the rights and freedoms of others."    [Back]

Note 169   Clayton and Tomlinson, The Law of Human Rights (2nd edn, 2009), vol i, paras 12.264–12.273.    [Back]

Note 170   "Article 4 (1) No one shall be liable to be tried or punished again in criminal proceedings under the jurisdiction of the same State for an offence for which he has already been finally acquitted or convicted in accordance with the law and penal procedure of that State. (2) The provisions of the preceding paragraph shall not prevent the reopening of the case in accordance with the law and penal procedure of the State concerned, if there is evidence of new or newly discovered facts, or if there has been a fundamental defect in the previous proceedings, which could affect the outcome of the case. (3) No derogation from this Article shall be made under Article 15 of the Convention."     [Back]

Note 171   That is, items of physical evidence.    [Back]


BAILII: Copyright Policy | Disclaimers | Privacy Policy | Feedback | Donate to BAILII
URL: http://www.bailii.org/scot/other/SLC/Report/2009/218.html