F & B, R v

LORD JUSTICE ELIAS: This is the judgment of the Court, to which each member has contributed.

We are faced with two applications for leave to appeal against certain rulings made by Gloster J in two distinct sets of preliminary proceedings heard on different occasions under section 9 of the Criminal Justice Act 1987. The case was referred to us by the Registrar on the basis that if leave were given, we would go on to hear the substantive appeals. Gloster J had refused leave. The Crown does not dispute that we have jurisdiction to hear all matters raised in these appeals pursuant to section 35 of the Criminal Justice Act 1987. We in fact have some doubts as to whether we do have jurisdiction to deal with the particular issue of delay. However, we have reached no conclusion on that point. We have heard no argument on it, and in view of our conclusions on the merits, we have not had to resolve it.

The background in bare outline is that both applicants were charged with one count of conspiracy to defraud contrary to common law in respect of the period 1 January 1998 to 30 June 2002. There was a third defendant tried at the same time, namely Nicholas Fraser, the younger brother of the applicant, Lincoln Fraser. (A fourth defendant had his case severed and was subsequently acquitted on the direction of the judge when the prosecution offered no evidence, and a fifth pleaded guilty at an early stage.)

There was a very lengthy fraud trial which took place from 31 March to 29 September 2008. After eight days of deliberations the jury was discharged. They failed to reach verdicts in respect of both applicants and notified the judge that they were deadlocked 6:4; it is not clear which way. However, Nicholas Fraser was acquitted.

A retrial was ordered with respect to the two applicants and this retrial is due to begin on 28 September, later this month that is. Hence the urgency of this, our reasoned judgment, today. An application was made to Gloster J by both defendants that the criminal proceedings should be stayed as an abuse of process. A number of different grounds were relied upon. The judge dismissed the application following four days of submissions in a ruling made on 8 April 2009, with reasons given on 27 April 2009. The applicants now appeal her ruling on most of the grounds she rejected.

By a further ruling made on 21 July 2009, Gloster J held that certain documents emanating from the acquitted defendant, Nicholas Fraser, were admissible in evidence. She rejected a further application from the Crown that certain passages from the evidence given by the two applicants at the first trial, in the form of LiveNote transcripts, should also be admitted. The Crown does not appeal that part of the ruling but the applicants appeal her finding with respect to the admissibility of the Nicholas Fraser documents.

Before considering these rulings we will summarise the facts and circumstances which gave rise to the prosecution of these defendants. We gratefully adopt the succinct summary given by the learned judge herself:

Leading counsel for each applicant, Mr Solley QC for Mr Jared Brook and Mr Smith QC for Mr Lincoln Fraser, initially submitted that the retrial would amount to an abuse of process on three grounds, namely delay; insufficiency of evidence; and uncertainty. In the course of submissions before the court, Mr Solley conceded that for reasons we give later, he could not properly contend before us that the retrial should be stayed because of insufficiency of evidence, but he did wish to pursue an admissibility argument which was inextricably related to that submission. So only two grounds relating to the stay application remain. The delay argument is that it would be unfairly prejudicial to the applicants for the retrial to be held so long after the alleged conspiracy. The uncertainty ground is a submission that the offence of conspiracy to defraud contrary to common law is so vague and imprecise that it infringes the requirements of Articles 5 and 7 of the European Convention on Human Rights, as well as the principles of domestic law.

The submission is that each of these grounds independently constitutes a reason why there should be a stay for abuse of process. A further ground, namely that the SFO had not followed its own procedures when deciding to pursue a retrial, was advanced before Gloster J but has not been renewed on appeal.

The legal principles relating to the circumstances when a trial may be stayed for abuse of process resulting from delay are now well established. It is accepted by all counsel that they are succinctly stated in Archbold (2009) para.4-66 in the following terms:

These common law principles are further supported by Article 6 of the European Convention which provides that a person charged with a criminal offence is entitled to a "fair and public hearing within a reasonable time". It is clearly established that this obligation is quite distinct from the guarantee of a fair trial: see Porter v Magill [2002] AC 357 HL. Accordingly, this obligation may be infringed even where a fair trial is possible. However, it does not follow that where this separate duty is breached there must be a stay of proceedings. On the contrary, that will only be necessary where a lesser sanction would be inadequate to vindicate a defendant's Convention right to a hearing within a reasonable time. The relevant jurisprudence, both domestic and Strasbourg, is carefully analysed by Lord Bingham giving the judgment of the Privy Council in Spiers (Procurator Fiscal) v Ruddy [2008] 1 AC 873. After considering a number of authorities, he concluded that it was only necessary to stay proceedings where either a fair trial was not possible or it would for some other compelling reason be unfair to try the defendant. Where neither was the case, the trial should not be discontinued but some other sanction would be appropriate:

As Lord Bingham's judgment also confirms, even where it would be possible fairly to try the case, there is a residual and important discretion to stay proceedings for abuse of process if for any compelling reason it would be unfair to try the defendant. The principle of abuse of process must protect a defendant from oppression or prejudice: see Connelly v DPP [1964] AC 1254 HL.

For the most part, the arguments which were addressed before us were the same as those which failed to persuade Gloster J. We are being asked to call into question her conclusion that it was not an abuse of process for this retrial to go ahead. It is important to emphasise that the question of whether a trial should be stayed for abuse of process is primarily one for the trial judge, having regard to such considerations as fairness to the defendant and the public interest. If the decision is one which could properly be reached by the trial judge then this court will not interfere. So the focus of this court must be on her decision.

The applicants advance a wide range of inter-related arguments as to why the delay here is so extensive that it would now be contrary to the public interest, and unfair to the defendants, for a retrial to be permitted. Some of these arguments relied upon failings by the prosecution, but it was also submitted that even independently of that, this was an exceptional case - indeed, Mr Solley submitted that it was virtually unique - where a retrial would be unfair and prejudicial. He emphasised that the delays were not, as is typically the case, the consequence of the defendants concealing their wrong doing, or becoming fugitives from justice. They were simply a consequence of the complex nature of the trial, exacerbated to some extent by certain failings by the prosecution.

The applicants point out - and we each of us see the force of this - that by the date of retrial it will be seven years since the offence was completed and over five years since the arrest. They observe that Lord Widgery in R v Saunders [1973] 58 Cr.App.R 248, noted that a retrial some three-and-a-half years after the original offence was committed was highly unusual. Here, the period is, of course, significantly greater.

The effect of the delay was, say the applicants, profound. Much of the evidence relied upon by the Crown was given by witnesses whose recollection of events for periods, in some cases more than ten years ago, would inevitably be very poor. Indeed, there were many occasions even at the trial where witnesses indicated that they could not now recall potentially relevant evidence. The defendants too would be similarly adversely affected by this difficulty. Added to all this was the obvious physical and emotional stress faced by these applicants and the fact that although on bail they were the subject of limitations on their freedom which further prejudiced their position. This did not affect the fairness of the trial but was material to the question whether it was fair for the trial to proceed.

The applicants further submit that there is significance in the fact that there are certain additional investigatory powers which are conferred on the Serious Fraud Office by section 52 of the Criminal Justice Act 1987, over and above those available to the Crown Prosecution Service and the Police. The purpose behind these powers, it is said, is to ensure that the prosecution have every opportunity to prepare their case thoroughly in advance of the trial.

The converse of that, so it is submitted, is that it should only be in exceptional circumstances that any retrial ought to be ordered in a serious fraud case. Indeed, it was submitted that there should be what is in effect a convention to that effect.

There were also certain unacceptable delays to be laid at the prosecution's door, which constituted a clear breach of the Article 6 duty. The history was that there was a criminal investigation which began when search warrants were executed in November 2002. The applicants were then arrested on 3 February 2004, but they were not charged until 13 June 2006.

They submitted that in accordance with the analysis iof Lord Bingham in Attorney General's Reference No 2 of 2001 , 2003 UK HL 68 at paras 26-27, the point at which the period of reasonable time should be determined for Article 6 purposes was when the defendants were "in a state of uncertainty about [their fate]". They submitted that this was November 2002.

The Crown submitted before Gloster J, that there were no special rules in cases of this kind. A fair trial could be held in 2008 - and no-one suggested at that stage that it could not - it could not be said that the delay of one further year would make any material difference. As far as the witnesses were concerned, there were witness statements as well as the LiveNote transcripts of evidence at the first trial which could assist any witness needing to assist their recall. Wider considerations, such as the strain of a further trial, were always potentially relevant in a case of this kind but they did not provide a compelling reason here why it would be unfair to conduct the retrial.

They denied that they had been responsible for any unreasonable delay. They submitted that the dictum of Lord Bingham in Attorney General's Reference No 2 of 2001 established that in a typical case (and there was no reason to treat this any differently) time would not run for the purposes of Article 6 until the defendant was formally charged or served with a summons. Lord Bingham stated in terms that the arrest will not ordinarily mark the beginning of the period, a fortiori nor will the beginning of investigations.

The case here had been progressed expeditiously from the dates of charge in June 2006. There had been certain delays but in part these were the product of actions by the applicants themselves. They had changed their counsel, for example. In all the circumstances there was no reason why the trial should not go ahead.

The judge accepted this submission. She took the view that the relevant date for the purposes of Article 6 was June 2006 and concluded that on that basis there had been no unreasonable delay. She noted that in determining that question consideration had to be given to the complexity of the case and the reasons for delay, this was required both by Convention jurisprudence: see Konig v Germany 1 EHRR 55, Porter v Magill [2004] 1 AC 379 at 396-402 per Lord Hope. The judge reiterated there had been no complaint about delay before the first trial and there was no unreasonable delay between the original trial and the retrial. She accepted that if indeed the period of unreasonable delay did begin at some stage before charge, whether the date of arrest or the date of criminal investigations, then there was a stronger case for saying there had indeed been unreasonable delay. She did not feel she had sufficient material to determine that question, but held, and this is important to the conclusion in any event, that even if there had been unreasonable delay resulting from a consideration of that period, it could be met by some sanction other than a stay of the proceedings. It was not necessary to discontinue the whole action in order to vindicate Article 6 rights. She was satisfied- and was in a particularly good position to assess this- that there would be no serious difficulties facing witnesses who wished to give evidence, in particular because of the material available to enable them to recall their evidence. She focused on the fact also that the fraud in this case was a very serious one involving losses to many people in excess of £250 million.

As we have emphasised, the question is whether this conclusion was open to the judge. The submissions on this point, if we may respectfully say so, came close to re-running those arguments before the judge. But it is not our job to second-guess her decision. We think the conclusion she reached was open to her, notwithstanding the time elapsed since the offence was committed. It is not possible, we think, to say that the decision to order a retrial demonstrated any error in approach or was outwith the range of decisions open to her. Gloster J had presided over the original trial and she was in a particularly good position to determine whether it would be possible for a fair trial to be conducted and to consider the impact of the delay on the recollections of witnesses. It is for the prosecution to decide whether to request a retrial and for the judge to ensure that if that order is made it will be fair to the defendants. Gloster J carefully analysed that question and concluded that it would be. In our judgment that was a cogent analysis. It discloses no error of law. Once it was determined that a fair trial could be held, then the other factors such as stress on the defendants provided no separate compelling reason why the proceedings should be stayed Accordingly, we refuse leave on this ground.

The submission here is that that Articles 5 and 7 of the Convention are infringed by the existence of this very general conspiracy offence. Arguably, this is not so much an argument that it would be an abuse of process for the retrial to proceed, but rather that there is no proper criminal offence for which the defendants are being charged - or at least that the conspiracy to defraud is not in this case such an offence. The effect of these Articles is that someone should only be found guilty of offences "prescribed by law".

The underlying principles here, as the ECHR pointed out in the seminal case of Sunday Times v United Kingdom (1979) 2 EHRR 245 para 49, is that the law must be accessible and that it must be formulated with sufficient precision to enable the citizen to regulate his conduct. As Lord Bingham pointed out in R v Rimmington [2006] 1 AC 459; [2005] UKHL 63, these principles are already well established at common law which is entirely consistent in this respect with the requirements of Convention law. His Lordship extracted two guiding principles, which he summarised as follows:

The applicants contend that these principles are infringed by the nature of the offence of conspiracy to defraud. The basis of the argument is that the definition of this common law offence is too vague and imprecise to satisfy Articles 5 and 7, or indeed the common law requirements. The defendant would not know in advance what conduct was forbidden and what was not. Many parties, including The Law Commission in its report on Fraud in 2002, various academics, judges and others have expressed concerns about the width of the offence. A particular objection, although only one of a number, is that since the offence relies exclusively on the jury being satisfied of dishonesty, the consequence is that leaving that issue to juries means that they may find that something is dishonest simply because it conflicts with their own perception of what is moral or right.

The judge held that there was sufficient certainty to satisfy the requirements of the common law and Convention jurisprudence. She noted that Parliament had, in fact, decided to retain this offence notwithstanding the criticism of The Law Commission when Parliament enacted the Fraud Act in 2006. Moreover, Judge LJ (as he then was), in R v Bright [2008] EWCA Crim 462 had recognised that in some cases counts of conspiracy to defraud are appropriate because of the magnitude of the defendant's criminality. As the judge noted, there are certain authorities which lend support to the conclusion that the offence does not infringe the principle of legal certainty required by Convention law. We shall consider some of the more relevant ones.

In Norris v Government of United States of America [2007] EWHC 71, Admin, the Divisional Court (Auld LJ and Field J) were concerned with the question whether prior to the enactment of the Enterprise Act 2002 a price fixing agreement per se, when entered into secretly, constituted a conspiracy to defraud at common law. The court held that it did. One of the arguments addressed to the Court was whether the offence of conspiracy to defraud was too uncertain. Auld LJ noted that there was a high Strasbourg threshold before domestic law would be regarded as so uncertain as to be incompatible with Article 7 and cited a number of cases supporting that proposition. Auld LJ concluded (para 98) that that threshold had not been met here. He held that the Ghosh test of dishonesty was sufficiently clear to satisfy the Convention requirements.

When the case went to the House of Lords, their Lordships upheld the appeal and held that the secret price fixing agreement was not indictable "in the absence of aggravating features such as fraud, common misrepresentation, violence, intimidation or inducement to breach of contract (see paragraph 17 of the consolidated Opinion). Here no such features had been identified. The mere fact that the agreement was secret could not amount to dishonesty. The House did consider the concept of certainty but only in the context of determining whether a particular price fixing agreement of the nature identified in that case was sufficiently well established at common law. They did not specifically address the wider argument which has been raised before us.

The decision in Norris was applied in the case of R v GG plc & Others [2008] UKHL 17. In that case the court held that the way in which the indictment had been framed, simply doing no more than alleging that the defendant had conspired to defraud the Secretary of State for Health by dishonestly fixing and maintaining the price and manipulating the supply of certain products, did not disclose anything potentially dishonest. The court noted, however, that there was in fact voluminous material in the Case Statement such as evidence of lies and positive deception which could perfectly properly have been relied upon in a charge of conspiracy to defraud. The court said this (para 18):

In our judgment, although the issue of Article 7 compatibility was not specifically before their Lordships, this decision strongly supports the conclusion that provided the elements of the unlawful means are properly identified, there is no infringement of Article 7.

In this context we think that valuable assistance can be gleaned from the decision of the Court of Final Appeal (Hong Kong) in the case of MO Yuk Ping v HK SAR [2007] HKCFA 52. The judgment was given by the distinguished Australian judge, Sir Anthony Mason, sitting as a non-permanent judge in the court. The issue there was whether the common law offence of conspiracy to defraud was consistent with requirements of Hong Kong Law which in this respect are very similar to the requirements imposed by UK and Strasbourg law. Sir Anthony Mason analysed the history of the offence of conspiracy to defraud and demonstrated how it had been extended beyond an intention to deceive to embrace any dishonest means. He concluded that there was no formulation which encapsulated precisely the nature of the offence, but he summarised what he considered to be the key elements in the following way (para 40):

His Lordship then considered whether the offence so defined was sufficiently clear so as to satisfy the requirements of certainty. He concluded that it was in the following terms:

We respectfully agree with that analysis. We see some force in the submission that, notwithstanding the decision of Norris in the Divisional Court, an allegation of conspiracy to defraud which relies merely on a generalised allegation of dishonesty, without identifying any specific unlawful means, might offend the need for certainty. But we do not think that this would be an appropriate formulation of the offence. In any event, the possibility that this may be so does not mean that the offence can never be charged even where the relevant criminal means are identified. As the Hong Kong Court makes clear, there needs to be an allegation of dishonest means of some kind. In our view Norris and GG are entirely consistent with that analysis.

Accordingly, in our view, at least where the conspiracy in issue is one where the allegation involves dishonest means and not merely a generalised allegation of dishonest conduct, this offence satisfies the requirements of the Convention and is sufficiently certain.

It seems to us that in this case the offence was analysed in this way at the first trial. The judge's summing up shows that the prosecution case as to fraud and dishonesty fell under three headings. The first was false and dishonest representations to investors. Here 'dishonest' must mean known by the defendants to be false. The second was the dishonest presentation of false accounts for the three months to July 1999 and 17½ months to September 2000. The third was dishonestly continuing to trade when they knew that there was no realistic possibility of the investors being repaid, but nonetheless they continued to accept investments. In the case of each heading it would follow that if the defendants had acted as alleged their conduct was dishonest in accordance with the Ghosh direction. On this analysis the real issue for the jury was whether the defendants had acted as the prosecution alleged under the three headings.

In short, the conspiracy relied upon was conduct which was itself unlawful - misrepresentations, false accounting and unlawful trading. Nobody, we consider, could have been in any doubt that such conduct was criminal and it had not only become apparent that it was criminal at the trial itself. The conspiracy was, in other words, what we may call an "dishonest means” conspiracy rather than one which simply relied upon generalised allegations of criminality. On the assumption that the retrial is conducted in a similar way and identifies the relevant unlawful means, we consider that there is no question at all of lack of clarity infringing Convention law.

This ground, relating to admissibility of evidence, proceeds principally on the implications of the acquittal of Nicholas Fraser. The applicants initially contended that it would be wrong for the court at the retrial to seek to adduce evidence which was inconsistent with Nicholas Fraser's acquittal. If the court took a proper view of the evidence and had regard only to that which could properly be adduced, then whilst it could not be said that there was no evidence against these applicants, it was plainly insufficient to sustain the conviction of either of them. In those circumstances it would be an abuse of process to allow this case to go to the jury.

There are therefore two elements to this submission. The first is that certain evidence is not admissible i.e that evidence which is inconsistent with the jury's acquittal of Nicholas Fraser. The second is that the effect of omitting that evidence is that insufficient is left to sustain a prosecution.

We think that notwithstanding some concerns expressed by the Crown, we can properly consider the first issue. Mr Solley conceded that we could not sensibly deal with the second. It seems to us that the judge, for perfectly good reasons, did not consider what would be the effect of the decision if the applicants' submissions on admissibility were correct; she did not think they were. We suspect that given that in many material respects the evidence with respect to them was different to that which was being relied upon against Nicholas Fraser, she would have found that there was sufficient evidence. Whether that be so or not, we are in no position at all to determine whether, absent the evidence the applicants wish to have excluded, there is sufficient remaining to sustain convictions. The most we can do at this stage is consider the admissibility issue as a self-standing issue, independently of the question of stay.

So we turn to that question. The heart of the submission is this: Mr Nicholas Fraser was acquitted by the jury. On the facts of this case - and Mr Solley emphasises that he is not seeking to lay down any more general principle - the only reasonable inference from that acquittal, given the way in which the case was presented by the Crown to the jury, is that they were not sure of Mr Nicholas Fraser's dishonesty. The essence of the case against him was that he was knowingly and dishonestly involved in the creation of false accounts.

It would now be unjust, say the applicants, for the prosecution to seek to adduce evidence which sought to establish any fact which could not sit consistently with the jury's finding. In particular, the Crown were seeking to keep open their option to contend before the second jury that Mr Nicholas Fraser was dishonest. Since the only proper inference from the acquittal is that the earlier jury was not satisfied of that, it would be abusive and wrong to allow that issue effectively to be re-run before a second jury.

True it is that Mr Nicholas Fraser is not a defendant in these proceedings and so it is not a case of double jeopardy, but there is an underlying principle that the trial should be fair and it would be unfair to allow a second jury to come to a conclusion on an important central feature of the case which had already effectively been determined by the jury at the first trial.

Mr Solley did have some difficulty in identifying precisely what the implications of this conclusion would be. In particular, at one stage he contended that it would be improper for the Crown to seek to cast doubt on the veracity of any of the matters which the Crown had advanced before the jury in their closing submissions as demonstrating the guilt of Mr Nicholas Fraser. In particular there were some 20 points identified by the Crown in their closing speech. A more limited submission was that at the very least it would be wrong to allow the prosecution to seek to adduce evidence at the retrial that the accounts produced by Nicholas Fraser, who after all was the Finance Director of the company, were dishonestly prepared. It would be open to the prosecution to say that they were false, and even to contend that the applicants knew this, notwithstanding that Nicholas Fraser did not, but they could not properly seek to ask the jury to infer the dishonest guilt of the two applicants from the alleged dishonesty of Nicholas.

Reliance is placed upon Article 6 of the Convention, and in particular the requirement, which has been implied by the ECHR, that reasons must be given for judicial decisions. Certain authorities, both domestic and in the ECHR, are to the effect that a verdict of a jury, even where the jury gives no reasons, can be consistent with Article 6 but only if it could be read with other material, so that in effect, one could infer the jury's reasons: See Hiro Balani v Spain [1994] 19 EHRR 565; R v Belgium Application 15957/90; and Saric v Denmark Application 31913/96. In accordance with this principle, it is said that the court should do its best to make sense of the jury's verdict so as to identify the reasons for the acquittal. If that is done, the only proper inference here was that the jury had rejected the Crown's submissions.

With respect to this aspect, we make two observations. First, these cases are concerned with convictions rather than acquittals. It will be relatively easy in most cases at least to discern the reasons for a conviction; the jury will have been sure that the elements of the offence were satisfied. That is not, however, the case with respect to an acquittal. There will often be any number of reasons why a defendant could have been acquitted, and no doubt generally the acquitted defendant is not unduly concerned what the reason was. Second, and in any event, whether or not the jury system could be said to infringe Article 6 (we do not say tat it would) because of a failure to enable reasons to be identified for an acquittal, that would not entitle a court to strain to infer reasons where no proper inference could be made.

The Crown contended that the fact of Nicholas Fraser's acquittal did not begin to establish that the jury had rejected all the matters identified by the Crown in closing submissions. As the judge had indicated, there were other possible explanations for the acquittal, including the possibility that Nicholas was dishonest but not a party to the conspiracy involving the other two defendants. The acquittal simply meant that the jury had not been sure that the offence of conspiracy to defraud had been established with respect to that defendant. It is illegitimate to go further and infer from that verdict that the jury must have been uncertain as to dishonesty.

In any event, the Crown submit that even if it were possible to infer that the first jury were not satisfied that Nicholas was dishonest, there was no rule of law prohibiting the Crown from seeking to establish otherwise at the retrial. It was not in fact their current intention to ask the jury to infer that the two applicants were guilty of producing dishonest accounts by relying on the dishonesty of Nicholas, nor indeed at this stage were they necessarily going to seek to establish that he was dishonest. But they reserved the right to do so. The jury on the retrial would be informed that Nicholas had been acquitted and it would be open to the applicants to seek to persuade the jury that this supported the view that he was honest. But the acquittal did not compel the prosecution to conduct the retrial on the assumption that he was. There was no issue estoppel operating in criminal law (see R v Humphrys [1977]AC 1) and provided there was no double jeopardy - and there plainly was not here since Nicholas was not even a defendant in the second trial - then there was no restriction on the evidence that could be adduced as a consequence of the acquittal.

Mr Barnes QC, counsel for the Crown, relied upon a number of cases in support of his submissions, but the principal one was R v Terry [2005] 3 WLR 379, and the authorities to which it referred.

The judge accepted the Crown's submission. She held both that it was not legitimate to infer from the acquittal that the jury must have had doubts whether Nicholas was dishonest, and she accepted that the authorities demonstrated in any event that the prosecution could reopen the issue of his dishonesty. Furthermore, she noted that it was an impossible task to identify precisely what evidential issues were inconsistent with the points relied upon by the Crown in their closing speech.

We agree with the judge. It is a necessary step in the applicant's argument that the only proper inference from the jury's acquittal was that they must have been unsure whether Nicholas was dishonest. The judge rejected that as the only proper inference. We confess that we have harboured some doubts about how realistic it is to conclude that the jury could have acquitted on grounds other than they were unsure whether Nicholas was dishonest, for example, because they were not sure that he was a party to the conspiracy. But that is plainly a logical possibility and provides an explanation for their verdict and we have concluded that the judge was in a far better position than we are to interpret the jury's verdict, or perhaps more accurately, to conclude that no valid inference could be drawn from it as to the reason for the acquittal.

In any event, we accept the Crown's submission that there are clear authorities, and in particular the most recent decision of Terry , supporting their submission that the evidence is in principle admissible. In Terry the defendant and others were charged with various offences of burglary, theft and handling stolen goods. The prosecution sought to rely upon certain conversations which had been covertly recorded by the Police installing a listening device in a car used by the defendant. A voice recognition expert was called by the Crown and she gave evidence that on the assumption that there was other evidence confirming the presence of the defendant when the recordings took place, in her view the voice was that of the defendant. The judge ruled that the expert's opinion in evidence on the four counts to which her evidence related was not admissible. The Crown offered no evidence on those counts and the jury at the direction of the judge entered verdicts of not guilty in relation to them.

During the trial on the remaining counts the defendant sought to pray in aid those four acquittals. His case was that he could adduce evidence of the acquittals as conclusive proof that he had not been in the car at the time of the excluded conversations. The judge accepted that the acquittals were potentially material and could be tendered to support his contention that the car was used by others at the time of the alleged incriminating conversations, but that he was not entitled to the direction he sought. The defendant was convicted of burglary, theft and handling stolen goods and he appealed on the ground that the earlier acquittals were not only admissible on the grounds of relevance, but should have been treated as positively conclusive evidence for all purposes that the defendant was innocent of the offences of which he had been acquitted.

The Court of Appeal (Auld LJ, Owen and Hedley JJ) dismissed the appeal. They rejected the contention that the acquittal was conclusive evidence of innocence. There was no question of double jeopardy here because the defendant was not convicted of the particular offences of which he had been acquitted.

It is instructive to consider the case of R v Hay (1983) 77 Cr.App.R 70 which was analysed by the Court of Appeal in Terry . The defendant in that case had made a single confession in which he had admitted to two unrelated charges, one of arson and one of burglary. The indictment was severed and in the arson trial he was found not guilty. His evidence had been that the Police had fabricated his confession and that he had an alibi. He sought to put in evidence in the burglary charge the jury's acquittal in order to support his submission that the confession could not be relied upon. The trial judge held that the evidence was inadmissible because it was irrelevant to the burglary charge.

This court allowed the appeal against conviction. O'Connor LJ, giving the judgment of the court, said this (page 75):

That judgment would, on the face of it, lend some support to the applicant's submissions, save that it is of course directed to a situation where the defendant himself rather than third parties was being prosecuted for the second time.

However, the Court of Appeal in Terry concluded that the decision should not be treated as binding. First, they noted that O'Connor LJ's observation went further than the defendant himself had sought. He merely wished to be able to adduce the acquittal in evidence, whereas the court held that he must be deemed to be positively not guilty of the earlier offence, and his confession should be treated by the second jury as untrue.

Furthermore, the court noted that it was inconsistent with certain other authorities. First, in R v Ollis [1900] 2 QB 758 the Court for Crown Cases Reserved had accepted that in a trial of obtaining money by false pretences on worthless cheques, the prosecution was entitled to adduce evidence which had been unsuccessfully relied upon in an earlier trial of a similar nature, notwithstanding that the defendant had been acquitted. The court clearly distinguished between the rule of trying a man twice for the same offence and rule relating to the admissibility of evidence relevant to his guilt of the offences charged in the second trial. The court held that there was no estoppel. Provided the evidence was relevant it could be adduced. Lord Russell of Killowen, CJ, who gave the leading judgment, stated in terms:

More recently in R v Z [2000] 2 AC 483, the House of Lords held that in a trial for rape where the issue was consent, the jury was entitled to hear the testimony of previous complainants who had alleged that they had been raped by the defendant even though he had been acquitted of charges of rape with respect to those complainants.

The evidence relied upon in that case was similar fact evidence and Mr Solley sought to limit the significance of the case by contending that it should be so restricted. However, Auld LJ in terms stated that there was no legal basis for restricting the principle to such situations. He said this (para 45):

These cases were concerned with the position of the defendant himself facing a retrial. Mr Solley submits that this is a ground for distinguishing them in his favour. We do not agree. If evidence can be adduced with respect to the defendant then it seems to us a fortiori that the evidence is admissible where it is being relied upon against other parties, providing of course that it is relevant - and no-one disputes that here. Such very limited exceptions as there are to that rule are not applicable here.

We would add that if the argument were correct then, as Gloster J pointed out, it would mean that in virtually no case could a co-conspirator be retried following the acquittal of another defendant alleged to be a party to the conspiracy, since in practice evidence against co-conspirators would overlap significantly.

It follows that, in our judgment, there is no reason in principle why the prosecution should not be allowed to adduce evidence establishing that Nicholas Fraser was dishonest, although the defence can of course seek to pray in aid the acquittals in order to contend otherwise. Of course, to the extent that the prosecution seek to rely upon evidence given at the first trial to establish any fact, the judge will need to be alert to the obligation to ensure that there is no unfairness to the applicant resulting from any partial or potentially misleading selection of the evidence that was before the court. Mr Barnes accepted that that was plainly correct.

Finally we turn to consider the specific disclosure application relating to documents emanating from Nicholas Fraser. In a file stored in the hard drive of Nicholas Fraser's laptop headed "Letter to Jarrod.doc" created on 25 May 2000 and modified on 2 June 2000, Nicholas Fraser set out his concerns about the historic and current position of the group. In particular, he stated:

A second document was recovered from the same source entitled "Appendix 3 - Financial position.doc". It was created on 14 September 2000. A copy of it was sent to an external tax consultant, Peter Vaines. It stated:

The Crown do not propose to call Nicholas Fraser as a witness at the retrial because they do not consider they can rely on his honesty.

The Crown applied to put in both documents under sections 115 and 117 of the Criminal Justice Act 2003. Gloster J acceded to both those applications. Both defendants now seek permission to challenge her decision.

The relevant parts of those two sections are as follows:

Mr Solley and Mr Smith made a number of submissions opposing the admission of the documents. For present purposes three are emphasised:

We readily accept that a statement in a business document tendered for the purpose of proving a matter stated in it is only admissible under section 117 if when the statement was made the person making it intended that, by virtue of reading it (or, perhaps, by having it read to him), the other person should be caused to believe the matter stated in it. Thus, as Mr Barnes accepted, a diary note or aide memoire, created only for the purpose of the person creating them, would not be admissible as a business document under section 117. Mr Solley accepts that it is not necessary that the document should in fact reach, let alone be read by, its intended recipient. Gloster J addressed this issue directly. She decided, correctly in our view, that it was not necessary that the document was physically sent or handed to Jared Brook. She concluded that the evidence, in particular the title of the document, was consistent with Nicholas Fraser originally intending to send a letter to him but then deciding not to do so. We understand her there to state her conclusion that it appeared to her (to use the language of the section) that Nicholas Fraser's purpose in making the statement was to cause Jared Brook to believe the matters stated in it by sending or showing it to him, but changed his mind. On that basis she was right to rule that the statement was in principle admissible as a business document. Her conclusion is supported by the terms of the document which are explicitly addressed to Jared Brook and which suggest that it should not be read like an aide memoire.

No issue of admissibility in principle arises with respect to the "Appendix 3" statement, because that was sent to Peter Vaines and it plainly falls within section 117.

We accept Mr Solley's submission that the letter to Jared.doc statement is not an ordinary business document of the kind which has traditionally been routinely admitted under this section and its statutory predecessors. It is not a document "compiled by persons who are disinterested", as this court put it in Horncastle [2009] EWCA Crim. 964, para 11. But if it falls within the terms of the express statutory provision, it is admissible, subject to the statutory possibility of exclusion under section 117(7) and section 126(1) of the 2003 Act and section 78 of the Police and Criminal Evidence Act 1984, notwithstanding that it is nor the usual kind of document caught by the section.

So the question is whether the draft letter falls within its terms or not. The judge expressly addressed the challenges to the reliability of the statement under section 117(7). She dealt specifically with the two principal relevant yard sticks: the contents of the statement and the source of the information contained in it. As to contents, she concluded, plainly correctly, that the concerns expressed by Nicholas Fraser about the financial position of the group were supported by oral and documentary evidence led by the Crown. As to the source of the information, she noted that in the document Nicholas Fraser complained about his limited knowledge of the CFBH deal and the rebuffing of his attempts to enquire into the financial condition of the group, but noted, also correctly, that the document appeared to show that in the recent past he had been given access to more financial information - which, as he noted, had enabled him to put together a balance sheet and profit and loss projections, which made the state of affairs apparent. We consider that she was entitled to reach those conclusions.

No separate submissions as to reliability was made under section 117(7) in relation to the "Appendix 3" document. This is unsurprising, as the comments about the financial history and position of the group, with the exception of the optimism expressed about its future prospects, was entirely consistent with the oral and documentary evidence adduced by the Crown.

The judge decided that the reliability of neither document was shown to be unreliable for the two purposes for which they were to be tendered, namely to demonstrate the financial position of the group and the use of the new investors' funds to meet old liabilities, and to demonstrate the state of mind and knowledge of the finance director, Nicholas Fraser.

The first purpose sits squarely within section 117. The second does not. Although the issue was not addressed in submissions to the judge, and nor was it considered in her judgment, the documents are not admissible for that second purpose under section 117. That purpose is not to prove the matters stated in the document, but to demonstrate the state of mind and knowledge of the author and, perhaps, by inference of his co-directors. The documents are therefore not hearsay with respect to that purpose. They are however real evidence of his state of mind and knowledge and so are admissible at common law. The prosecution strictly therefore did not need to rely upon the terms of the statute for that purpose, so it seems to us.

Gloster J considered and rejected submissions that she should exclude the documents under section 126(1) of the 2003 Act and/or under section 78 of the Police and Criminal Evidence Act 1984. Both provisions permit statements to be excluded, but for different reasons. Section 126(1)(b) is primarily directed at the orderly management of the case: the court can refuse to admit a statement in evidence if "satisfied that the case for excluding the statement, taking account of the danger that to admit it would result in undue waste of time, substantially outweighs the case for admitting it, taking account of the value of the evidence". Section 78 is concerned with the impact of the admission of the statement on the fairness of the proceedings. It may be excluded if it appears to the Court that the admission of the evidence would have such an adverse effect on the fairness of the trial that the Court ought not to admit it. Gloster J concluded that the documents were of value because they dealt with issues which lay at the core of the case and that their value was not undermined by the absence of Nicholas Fraser. She also noted that time was not wasted in the first trial in relation to them. Again, we note that it is not for us to make this determination, but to consider whether or not she was entitled to reach those conclusions. We think that she was. She also concluded that the admission of the documents would not have such an adverse effect on the fairness of the proceedings that they ought not to be admitted, notwithstanding the fact that the Crown did not intend to call Nicholas Fraser to prove them. She said that it would be unreasonable and unrealistic to require the Crown to do so. Again, we think that she was entitled to come to that conclusion. Accordingly, we conclude that her findings on these matters are unimpeachable.

We add this. Nicholas Fraser gave an explanation for the documents and for his accounting treatment of the matters canvassed in them and it may be difficult for the defendant to adduce that evidence without calling him. It may be that they will decide to do just that. Alternatively, to avoid the expenditure of much time and effort on what may be a secondary issue in the retrial of Jared Brook and Lincoln Fraser, the parties might consider the making of an admission by the Crown as to the general nature of the explanation proffered by Nicholas Fraser with respect to these matters. The Crown would not of course be required to admit the truth of the explanation or the propriety of the accounting treatment. However, the concern expressed by Mr Solley and Mr Smith was that if the documents are admitted without reference to Nicholas Fraser's explanation, an unfairly unbalanced picture may be presented. To the extent that this is so, it seems to us that it could be substantially allayed if the approach we have suggested were adopted. Whether or not such a step is taken is for the parties to consider.

For these various reasons we dismiss each of these applications and refuse leave in respect of them.

Finally we pay tribute to the quality of the two judgments of Gloster J, which are the subject of this appeal.