England and Wales Court of Appeal (Criminal Division) Decisions

IN THE SUPREME COURT OF JUDICATURE
COURT OF APPEAL (CRIMINAL DIVISION)
ON APPEAL FROM THE CROWN COURT AT MIDDLESEX GUILDHALL
HHJ Smith

		Royal Courts of Justice Strand, London, WC2A 2LL
		16 October 2003

Lord Justice Rix:

1. These are the appeals of Rafiq Petkar and Martin Farquhar who were convicted on 17 June 2002 in the Crown Court at Middlesex Guildhall before HH Judge Smith and a jury, at a re-trial, by majority verdicts of 11:1 on two counts of theft. On 28 June 2002 both were sentenced to 5 years imprisonment on each count concurrent. A third count against Petkar of concealing or transferring the proceeds of criminal conduct was left on the file.
2. Petkar appeals against conviction with the leave of the full court and renews his application for leave to appeal against sentence. Farquhar's renewed application for leave to appeal against conviction came before us on 20 June 2003 at the same time as Petkar's appeal. Since that renewed application raised in part the same ground as that on which Petkar had already been given leave to appeal, we granted leave to Farquhar for that (and two other) grounds and were compelled to adjourn the hearing of Petkar's appeal part heard so that what had then become two linked appeals could be taken together. Fortunately, this court could be reconstituted without too much delay and we heard the two appeals together on 25 July 2003. We then gave leave for a further ground of appeal to be raised by both appellants, arising from the recent decision of this court in R v. Jones and Jenkins [2003] EWCA Crim 1966 (6 June 2003). At the close of the hearing we announced our decision that both appeals were dismissed with reasons to be reserved. We then proceeded to give Petkar leave to appeal against sentence and allowed that appeal, substituting a sentence of 4 years and 6 months for the original 5 years. Our reasons for these decisions are now contained in this judgment.

The structure of the trial

3. Petkar and Farquhar were both employees of the Industrial Bank of Japan ("IBJ") in the City of London until October 1998 in the case of Petkar and December 1998 in the case of Farquhar. At that time Petkar was 37 years old and Farquhar 28. On 27 November 1998 Farquhar transferred £420,000 from IBJ to an account at Barclays Bank in the name of Graceland Investments Limited, of which Petkar was the sole director. Barclays sent the money back to IBJ's account with the Midland Bank on 1 December, but while that was in train and before that had happened, on 30 November Farquhar transferred a further sum of £420,000 from IBJ this time to a personal account in the name of Petkar at the Halifax.
4. Farquhar did not dispute that he had made both transfers, nor did Petkar deny that he had received them. However Petkar's case was that he believed that the money belonged to Farquhar himself and came from Farquhar's own bank, the Jyske Bank, and had been transferred to him to invest on Farquhar's behalf. Farquhar's case, on the other hand, was that he knew the transfers were dishonest but that he acted under the duress of Petkar and his associates. Theirs was thus a full cut-throat defence, for each defendant's evidence, if believed, was deadly to the other; but the failure of any one defence did not automatically put the co-defendant in peril. Farquhar's defence of duress could fail absolutely without harming Petkar in the slightest. Similarly Petkar's defence could fail without impinging on Farquhar's case of duress.
5. Following the first transfer to Graceland's account there was a same-day transfer over to Petkar's personal account at Barclays. Following the second transfer, a total of £310,000 was taken out in cheques or drafts or electronic transfers on 1 December 1998, much of that in payments to gambling casinos. In the event the only sums recovered following Petkar's arrest was a balance of £21,526 in Petkar's Halifax account and a further sum of £91,762 to the credit of Petkar's trading account with Refco Overseas, a broker based in Bermuda: a total of £113,288. That £91,762 with Refco was what remained of a total of £125,000 transferred by Petkar to Refco: the balance of £33,238 had gone to fund margin losses previously incurred by Petkar on his own trading. What had happened to the remaining £273,474? Petkar's evidence was that £40,330 had been spent on himself, either in settlement of outstanding bills or in fresh spending: this sum could be substantiated by reference to documents. He also accepted a certain amount of cash spending. All in all, he accepted that £198,000 of the £420,000 had been either spent on himself or transferred to Refco, but he maintained that the balance of £222,000 had been given back to Farquhar, at his request, in cash, in stages between 3 and 13 December 1998. These transfers were not documented. Two cash withdrawals from the Halifax account of £18,000 on 3 December and £24,000 on 9 December were both said by Petkar to have been handed over to Farquhar as part of this process.
6. That was the evidence at trial. At the time of his arrest and first interview on 31 December 1998, however, Petkar told the police, when asked what had happened to the money, that it had all been returned to Farquhar. He explained that the money had been sent to him by Farquhar from Jyske to trade with, but that he had returned it all. He would have profited from the trades generated had the matter proceeded, but as it was all the money had been returned. He was also asked if he had had a written contract with Farquhar concerning the terms on which he was to deal with the money, and he said that he did not.
7. A fortnight later, however, on 15 January 1999, Petkar contacted the police to inform them that not all of the money had gone back to Farquhar. He then gave them the essentials of the information we have set out above and faxed them a copy of the statement of his Halifax account. The information that he then provided to the police remained his case at trial. The Crown did not dispute that some £222,000 or approximately half of the £420,000 had been provided in cash to Farquhar. Thus at trial Petkar accepted that what he had told the police originally about returning all the money to Farquhar was a lie. Farquhar denied receiving anything, and there was no direct evidence, other than Petkar's, that he did.
8. Petkar also accepted that he had lied in telling the police on the day of his arrest that he had no contract with Farquhar. In fact a document dated 30 November 1998 and purporting to be signed by Farquhar (there was ambivalent evidence at trial from two handwriting experts called by the defendants respectively as to whether the signatures were genuine or not) was already at that time in the hands of Petkar's solicitor, Mr Murrell, and had been since 21 December. Mr Murrell was to give evidence that the contract came to him in connection with a dispute between Petkar and Barclays over the return of the first transfer and Petkar's subsequent closure of all his Barclays accounts. That contract was not, however, disclosed to the prosecution until April 2000.
9. Farquhar, however, denied receiving any of the money. He had tendered his resignation to IBJ on 27 November 1998, the date of the first transfer, but had been asked to work out his notice. His last day at IBJ was 9 December. He left for Ireland on 12 December and arrived unannounced at the home of his girlfriend, Karen Sweeney, in Dublin on 13 December. He was later to say that he left to escape the attention of Petkar. He did not return to England until April 2001, nearly two and a half years later, after learning, as he said, from a friend who had watched the Irish equivalent of "Crime Watch" that he was wanted in connection with "something about London or the UK" but he did not know what. At that time Petkar was shortly due to stand trial.
10. Following his return he contacted the police through his solicitor and was interviewed. He began by denying any knowledge whatsoever of the two transfers of £420,000. (They had been made using the passwords of colleagues at IBJ.) Indeed, he claimed not to have known Petkar other than through limited contact within the work setting and denied any friendship or association with him. He denied asking Petkar to invest money on his behalf. He denied receiving any money from Petkar. When, however, the police informed him that they held transcripts of phone calls between him and Petkar and wished to question him about them, he declined on the advice of his solicitor to answer any further questions. Thus he made no admissions and said nothing about duress at that time.
11. His defence of duress was first exposed at a plea and directions hearing on 13 June 2001.
12. On 15 July 2001 there was an alleged incident at his home, where pellets holes were found in his front door. The next-door neighbour had heard nothing, but the next day a cartridge case was found. The police attended. On 25 September 2001, the night before he was due to give evidence at the first trial – which had been adjourned to accommodate him being prosecuted together with Farquhar – there was another alleged incident involving Farquhar and his car: three discharged rounds were found in the road, one bullet was found in the car's head rest, another bullet had gone through the head rest, and a third bullet mark was found in the road. The police again attended. As a result the first trial had to be aborted: hence the need for a retrial. There followed a police investigation, conducted independently of the police force involved in the prosecution, into whether the shootings were the responsibility of Petkar, as Farquhar alleged, or whether they were arranged by Farquhar himself in an attempt to pervert the course of justice. The investigation proved inconclusive and led to no prosecutions. At the retrial the Crown therefore considered itself required to take an entirely neutral attitude to these events, and cross-examined neither defendant about them. However, the officer in charge of the investigation, DS Beverley Mills, was tendered for cross-examination and each defendant gave evidence and was cross-examined on behalf of the other about these incidents; and in all other respects the Crown clearly set before the jury its evidence that Farquhar had been a willing actor not driven by duress.
13. Among that evidence were the tape recordings and transcripts of telephone conversations between the two defendants in the period between 26 November and 5 December 1998. Their voices were identified. Thus on 27 November at 1631 Farquhar is talking to a friend by the name of Mike and says: "Going, everything is gone to plan, mate…I just resigned…There's a lot going on, I'll tell you when I see you." When asked whether he had another job, Farquhar replied: "Yea, and some other things as well." 27 November was the day of the first transfer and Farquhar's resignation. On 30 November, the day of the second transfer, Petkar telephones Farquhar at 1054 and says: "Go into another room and call me back. I have a small problem, they've returned the 420 back to Midlands [IBJ's bankers]." At 1140 the same day Petkar rings Farquhar again and says: "I just spoke to Barclays again this morning, just a few minutes ago, and they have returned the funds this morning, so they definitely will show up this morning or this afternoon or during the day some time." Farquhar replied: "Yea so you still want me to make them do the thing again today…" At 1447 Farquhar called Midland Bank, asked if the £420,000 paid to Barclays the previous Friday was back into the IBJ account with the Midland and gave new instructions for the second transfer. On 1 December at 0906 Farquhar rang the Midland to chase whether the first transfer had come back. At 1016, during a conversation between Petkar and Farquhar, there is a reference to "everything is fine" and "you know our money".
14. We will refer to other aspects of the evidence below.
15. The main ground of appeal in the case of both appellants relates to the judge's directions pursuant to section 34 of the Criminal Justice and Public Order Act 1994. At the conclusion of the evidence and prior to final speeches during a discussion on the law the judge indicated that he intended to give the jury a Lucas direction on lies and a section 34 direction in respect of both defendants. There was no objection at that stage. During final speeches the Crown, then as now represented by Mr Milne, did not deal with the question of adverse inferences; but Mr Roche, then as now counsel on behalf of Petkar, did, giving them reasons why they should not hold Petkar's lies in interview about the existence of the contract and the return of all the money against him. It is not clear what Farquhar's counsel, then as now Mr Clarke, said on this subject in his final speech, but we will assume, with some hesitation, that he took the same line. On 14 June 2002, after final speeches but immediately before the summing up, Mr Roche on behalf of Petkar addressed the judge to the effect that he should direct the jury not to draw adverse inferences, alternatively to be precise as to the adverse inferences which might be in play. The judge rejected the first submission, and said he would be guided by the model direction set out in Archbold which Mr Milne drew to his attention. In the course of discussion, however, the judge had expressed concern about the multiplication of merely possible inferences. He said:

"I think that as long as I tell the jury the framework of law and remind them of the salient facts, I don't think it's desirable for me to suggest a lot of inferences which might not be in their mind anyway."

The grounds of appeal

16. Two grounds of appeal are shared by both appellants. The first, which I shall call ground 1, is, as we have just indicated, that the judge's section 34 direction was defective. Five complaints are made:

(i) that the judge failed clearly to identify the relevant inferences which the jury might draw;

(ii) that the judge failed to warn the jury that they should not convict wholly or mainly on the strength of an adverse inference;

(iii) that the judge failed to direct the jury that they could only draw an adverse inference if satisfied that the prosecution had established a prima facie case;

(iv) that the judge failed to direct the jury that they could only draw an adverse inference if they concluded that the only sensible explanation for the defendant's failure to mention the salient fact was that he had no answer at the time or none that would stand up to scrutiny; and

(v) the judge failed to remind the jury of the explanations given by the appellant for the appellant's failure to mention facts.

17. In all, it was submitted on behalf of both appellants that these defects either singly or in combination fundamentally flawed the judge's section 34 direction.
18. The second ground shared by both appellants ("ground 2") is that the judge failed to give the jury an appropriately structured, or indeed any, warning about the danger of relying on the evidence of each defendant as against the other: see Jones and Jenkins.
19. The other two grounds concern only Farquhar. One relates to the way in which the judge introduced in his summing up the evidence which Farquhar's grandmother, Mrs Winifred Wells, gave by way of statement admitted, following her death, under section 23 of the Criminal Justice Act 1988. The judge said that her statement "is not accepted and has not been tested in court". Mr Clarke submits that the judge was wrong to say that the statement was not accepted in circumstances where, although it was not accepted by Petkar, the Crown was merely neutral ("ground 3").
20. Finally, Farquhar relies on some new evidence of Petkar's accounts which only emerged after conviction in the course of confiscation order proceedings against him. A previously undiscovered personal account of Petkar came to light at the Royal Bank of Scotland ("RBS"), Ealing branch. Investigation showed that between October 2000 and June 2002 a total of some £177,000 had passed through this account. Mr Clarke submitted that these sums could explain the disappearance from sight of (at any rate a large part of) the £222,000 which Petkar said he had handed over to Farquhar in December 1998; and that if the jury had known of these sums they might have been more willing to accept Farquhar's case that he had received none of the stolen money ("ground 4").
21. Before we turn to each of these four grounds in turn, we need to say something further about the evidence given at trial.

The prosecution evidence

22. Witnesses from the IBJ gave evidence about the two defendants and the bank's operations. Petkar had been employed as a consultant from 1996 to 31 October 1998, when his contract expired, working on complex derivative problems. Farquhar worked in a different department, known as the cash management team, which was concerned with backroom preparation of accounting journals and matching of payments. He had commenced work with the bank on 6 July 1998 as a temporary, became a permanent member of staff shortly thereafter, and resigned, as we have said, on 27 November 1998. He was regarded as intelligent and hard working. He was not authorised to make payments.
23. A witness from Barclays Bank, Mr Sones, gave evidence about the first transfer and the circumstances in which that was returned. Barclays was suspicious about it, and Mr Sones spoke to Petkar requesting its return. By the time that Mr Sones had been advised that it could not be refused, Petkar said he had made alternative arrangements and wished to close his Barclays accounts, which he did on 30 November. In the circumstances all the arrangements which Petkar had made to deal with the money had to be reversed or alternatively were not acted on. Among those arrangements were instructions dated 30 November to make the following transfers: £100,000 to Petkar's account with Citibank; £100,000 to Petkar's account with the Halifax; £150,000 to Refco for "margin requirements"; and £26,000 in repayment of two loans. A critical question in the light of Petkar's defence was whether there was any evidence that he knew that the first transfer had come from IBJ and not Jyske. Mr Sones said he was aware that the money had come from IBJ but was unable to say whether that had been discussed with Petkar. However, a letter generated by these transactions, which also involved sending a cheque to Petkar, referred to IBJ. Petkar accepted receiving the cheque, but not the letter. The letter was not found at Petkar's home; but Mr Sones was confident that he had sent it with the cheque: it was the letter which carried Petkar's address in a window envelope. A Barclays bank statement for the Graceland account, dated 2 December, detailing the transfer as coming from IBJ was found at Petkar's home: but Petkar said he would have filed it without even looking at it.
24. There was also evidence about Petkar's visits to casinos. These had increased in early December, and on one occasion he had Farquhar with him as his guest. And on 3 December he had bought himself a Mercedes and paid £4,000 as a deposit, using the transferred funds to do so among other use of those funds for his spending, and debiting his Halifax account for future instalments due in respect of finance for the car.
25. Ms Sweeney's statement was also admitted under section 23 of the 1988 Act, because she had refused to come over from Ireland to give oral evidence. She said that Farquhar had turned up on 13 December 1998 and stayed about a fortnight. She expressed surprise that he had left his job at IBJ, and he said that he had a claim, but she was not sure what that was a reference to. He paid six months rent in advance for an apartment and was flashing money around. He suggested she should tell her mother that if the police came looking for him, she did not know him.
26. There was also evidence that in Ireland Farquhar opened a TSB account in the name of Mark Holland, assisted by a false job reference, a copy of a non-existent account in Rotterdam, a false accountant's report and a copy of a false passport. He then changed his name on 25 February 2000 by deed pole to Ferguson, using an accommodation address in Croydon and used that name in forming a publishing company. He subsequently reverted to the name of Mark Holland in buying a house on 29 January 2001.

Petkar's evidence

27. Petkar said that he had worked for some years in trading in financial markets. He had set up two companies of his own, one (Gracelands) for trading in derivatives and the other (Magenta Forex) for trading in foreign exchange. He had dealings with Refco. He had got to know Farquhar at IBJ. They talked about trading: Farquhar was interested and wanted to become a dealer. Farqhuar told him that he had money "outside" which he wanted Petkar to invest for him. Petkar gave him details of his bank accounts. A week before 27 November 1998 Farquhar phoned him (Petkar had left IBJ by then) to say that he was going to transfer £420,000 of his own money to him, from Jyske. He was pressed on how he could have thought that a young man in a relatively junior position at IBJ had so much money of his own: he replied that Farquhar had told him that he was just killing time at the bank and that he believed him.
28. As for the contract, Petkar said that was signed by Farquhar on 30 November in a bar in Piccadilly. Its termination date reflected the fact that the money would be needed for at least a year.
29. He was asked why the first transfer to Graceland had been transferred immediately to his personal account. His answer was that it was much easier to transfer money that is one's own rather than a company's. He was asked to explain the payment instructions generated by the first transfer: he said that having missed a projected deal because of the trouble with Barclays, he was going to use the money himself, and would repay Farquhar "from some fund or other". Farquhar would probably have objected, but would have accepted it after he had explained how it would come back from a future trade. "I would have repaid every penny. I had the whole thing on a spreadsheet in my computer, which the police took, and now it is no longer accessible."
30. He explained the large amounts of money paid to casinos out of the second transfer in the form of drafts as a device to enhance his credit, especially with potential clients: he denied that it was an attempt to launder the money.
31. He said that large amounts of cash had been paid to Farquhar. Despite the request to invest the £420,000 on Farquhar's behalf, Farquhar had within a day or two of the second transfer requested him to return £100,000 in cash. The first tranche was handed over on 3 December, the same day he had withdrawn £18,000 from his account in cash. £143,000 of cash was generated from the casinos. He met Farquhar several times at night, mostly at his own flat, to hand the cash over, in all £222,000.
32. He was asked about a letter he had written to the Financial Services Authority (FSA) on 15 December, in which he said that he had only one client (a Mr Zimmerman). He said nothing about Farquhar. He admitted that was a lie.
33. On 30 December he wrote another letter, this time to the Halifax, in which he said that the £420,000 was a short-term loan from Farquhar, since repaid. He admitted that was also a lie.
34. It was only on 31 December that he discovered that the money had come from IBJ. That morning, after finding that his Halifax account had been frozen, he phoned head office and learned that that had been done at the request of IBJ. So he spoke to IBJ, who told him that the money had come from it. He spoke to a senior manager, Kevin Merry, at IBJ several times that day. He told him that he had returned all the money to Farquhar and that he had nothing to do with it. At that very time he still had over £100,000 of IBJ's money. On the same day he was arrested and interviewed by the police and also told them that he had returned all the money to Farquhar, and that he had no contract with him, only a verbal arrangement. All this were also lies.
35. Why had he lied? The judge summed up his explanations as follows (at 49/50):

"The reason I was saying that I had none of the money was I was trying to distance myself from the situation. I lied at that time but everything I have said in court was the truth…

"I was telling [IBJ] that I was not a party to this theft and I was going to help them to recover the money. I suppose I would have gradually told them that I had some of it with me or available. Kevin Merry, however, never did call me in. I asked him what I should do and he [said]: Sit tight where you are."

"I lied during the interviews. I did them without a solicitor when I said that all of the – money had gone to Martin Farquhar. I said that because that was the same that I said to [IBJ]. So I continued with that lie.

"The reason I told the police I had no written contract over the investment was because of my problems with the [FSA]. I was frightened and I did not want to show anyone the contract."

36. As for Farquhar's defence of duress, he denied it. He denied having anything to do with the holes in Farquhar's front door or with the car. He was never called for a police interview in respect of these incidents.

Farquhar's evidence

37. Farquhar said he had always worked in financial markets since leaving school, and wanted to become a dealer in shares and options. He got to know Petkar, was impressed by the Reuters screen that he had on his desk, and let him know of his ambition to become a trader. Petkar said he would put in a good word for him. They visited bars in the West End together, and went to casinos some three or four times. One night Petkar introduced him, albeit not by name, to a couple of black guys, who he thought were called Gary and Ritchie. He told him they were proper criminals, and had been involved in armed robbery. The week before 27 November Petkar had invited him to his flat. Gary and Ritchie were there.
38. Petkar led the conversation, talking about an investment scheme. After a while Gary interrupted: "Fucking tell him!" At this Petkar said that he wanted him to transfer money from IBJ for a few days, when it would be returned. He refused, not sure whether it was some kind of personality test. Gary took out a knife and started chopping cocaine, then waved the knife in front of him, threatening him if he did not do what was asked. Petkar explained that it would only be for a few days, and it was his job to tidy up the paperwork afterwards. Before he left, Gary grabbed him by the hair and threatened him again. A few days later he saw the two men outside his home, where he lived with his 82 year old grandmother, Mrs Wells, and was again threatened.
39. All the details of the two transfers were given to him by Petkar, albeit the decision to use his colleague's password was his own idea: "I did not want my name all over the payments." He resigned not out of fear of Petkar, but as a way of getting out of the problem. He was pleased to learn that the first transfer had failed. However, Petkar spoke to him on his mobile, ranting and raving, to get him to make the second transfer and threatening to send "the boys" to kick shit out of his nan. As for the transcribed phone-calls, Petkar could make himself sound nice, and he, Farquhar could sound brave.
40. He went to Ireland to get away from Petkar, but also as the one place "where you could find me". The false names he used there thus had nothing to do with hiding from the police: to get a house it was necessary to use a "credit package" in the name of Mark Holland, and the change to Ferguson was because he did not like the name Farquhar. He never told Ms Sweeney to tell her mother that she did not know me.
41. As for his police interview on his return to England, he merely wished to say as little as possible, to be non-committal.

Ground one: the section 34 direction

42. Section 34 provides:

"(1) Where in any proceedings against a person for an offence, evidence is given that the accused –

"(a) at any time before he was charged with the offence, on being questioned under caution by a constable trying to discover whether or by whom the offence had been committed, failed to mention any fact relied on in his defence in those proceedings; or

(b) on being charged with the offence or officially informed that he might be prosecuted for it, failed to mention any such fact,

being a fact which in the circumstances existing at the time the accused could reasonably have been expected to mention when so questioned, charged or informed, as the case may be, sub-section (2) below applies.

"(2) Where this sub-section applies – …

(d) the court or jury, in determining whether the accused is guilty of the offence charged,

may draw such inferences from the failure as appear proper."

43. This provision has given rise, in the light of authority, to a model direction recommended by the Judicial Studies Board of some elaboration, which is set out in Archbold, 2003, at para 15-334 as follows:

"1. Before his interview(s) the defendant was cautioned. He was first told that he need not say anything. It was therefore his right to remain silent. However, he was also told that it might harm his defence if he did not mention when questioned something which he later relied on in court; and that anything he might say might be given in evidence.

"2. As part of his defence, the defendant has relied upon (here specify the facts to which the direction applies). But [the prosecution say/he admits] that he failed to mention these facts when he was interviewed about the offence(s). [If you are sure that this is so, this/This] failure may count against him. This is because you may draw the conclusion from his failure that he [had no answer then/had no answer that he then believed would stand up to scrutiny/ has since invented his account/has since tailored his account to fit the prosecution's case/(here refer to any other reasonable inferences contended for)]. If you do draw that conclusion, you must not convict him wholly or mainly on the strength of it; but you may take it into account as some additional support for the prosecution's case and when deciding whether his [evidence/case] about these facts is true.

"3. However, you may draw such a conclusion against him only if you think it is a fair and proper conclusion, and you are satisfied about three things: first, that when he was interviewed he could reasonably have been expected to mention the facts on which he now relies; second, that the only sensible explanation for his failure to do so is that he had no answer at the time or none that would stand up to scrutiny; third, that apart from his failure to mention those facts, the prosecution's case against him is so strong that it clearly calls for an answer from him.

"4. (Add, if appropriate:) The defence invite you not to draw any conclusion from the defendant's silence, on the basis of the following evidence (here set out the evidence). If you [accept this evidence and] think this amounts to a reason why you should not draw any conclusion from his silence, do not do so. Otherwise, subject to what I have said, you may do so."

44. The fifth paragraph deals with an alternative or additional paragraph to deal with the situation where legal advice to remain silent has been relied upon, as it was in the case of Farquhar. In the present case, however, no complaint is made on behalf of Farquhar about the judge's direction from this point of view, although it was much briefer than the model direction.
45. The judge, following a direction on lies which Petkar had sought to make in itself the subject matter of a ground of appeal, but on which he did not obtain leave to appeal, gave a section 34 direction in the following terms:

"The next thing is what happens if a person is being questioned by the police, the modified right of silence, so to speak, we had for the last five or six years. If a person is proceeded against for an offence, as both defendants were, and when they are questioned by the police if one or either of them fails to mention a fact relied on in his defence in the trial: if it is something which in the circumstances at the time you take the view he could reasonably have been expected to mention under questioning, then, you may draw inferences from the fact that he did not mention it.

"And I do not want to be specific here but such inferences where he had not thought it up, he had not decided on his defence, he had not decided whether to make use of that particular thing, it is entirely a matter for you.

"Examples here, and they are only examples, the fact that Mr Petkar failed to mention the contract, although it seems that the solicitor had a copy or that he had not repaid or he had not passed on the whole of the money to Mr Farquhar.

"In Mr Farquhar's case, he did not mention the threat of the two men, Ritchie and Gary, at all. Said he knew nothing about the transactions. Those are examples of how this might apply here.

"It does not affect the right of silence in the sense that nobody has to answer questions and the prosecution must still prove the case against them and they have got to have something by way of evidence, other than merely…that the defendant did not mention part of his defence. And it is for you to decide whether it is right to draw any inference, you do not have to."

46. It will be straightway observed that the judge did not in that direction refer to the matters of explanation relied on by the defendants for their silence or lies, as per para 4 of the JSB model. He did, however, refer to them elsewhere, in summing up each defendant's evidence as a whole (see paras 35 and 41 above). It is submitted on behalf of the appellants that that is not good enough.
47. We have been referred to a number of authorities about the section 34 direction. Thus in R v. Condron and Condron [1997] 1 Cr App R 185 this court said at 195A that it was desirable that a direction should be given along the lines previously indicated in R v. Cowan Gayle and Riccardi [1996] QB 373 at 381 as necessary in a direction under section 35 of the same Act concerning a defendant's silence at trial. Lord Taylor of Gosforth CJ had there identified five "essentials":

"1. The judge will have told the jury that the burden of proof remains upon the prosecution throughout and what the required standard is.

2. It is necessary for the judge to make clear to the jury that the defendant is entitled to remain silent. That is his right and his choice…

3. An inference from failure [to give evidence/to mention a fact] cannot on its own prove guilt. That is expressly stated in section 38(3) of the Act.

4. Therefore, the jury must be satisfied that the prosecution have established a case to answer before drawing any inferences from silence…

5. If, despite any evidence relied upon to explain his silence or in the absence of any such evidence, the jury conclude the silence can only sensibly be attributed to the defendant's having no answer or none that would stand up to [cross-examination/scrutiny], they may draw an adverse inference."

And in R v. Argent [1997] 2 Cr App R 27 at 32/33 Lord Bingham of Cornhill CJ analysed section 34 as requiring six formal conditions, of which the fifth and sixth are relevant to mention: namely (5) that the alleged failure by the defendant must be to mention any fact relied on in his defence; and (6) that the failure is to mention a fact which in the circumstances existing at the time the accused could reasonably have been expected to mention. Those two matters are of course taken directly from the language of the statute.

48. In the light of Petkar's 15 January 1999 disclosures to the police relating to the disposition of the £420,000 (correcting his earlier lie at interview that all the money had been returned to Farquhar) Mr Roche on behalf of Petkar drew our attention to the relevance of the following remark of Stuart-Smith LJ giving this court's judgment in Condron at 197E:

"Moreover, it is always open to an accused person who has failed to mention some important fact at interview, to communicate it to the police at any time before trial; but unless it is done promptly, it is unlikely to rebut any inference which might otherwise be drawn."

49. In R v. Birchall [1999] Crim LR 311, a case on section 35, the judge had failed in his direction to remind the jury that they had to find a case to answer before drawing any adverse inference from the defendant's silence at trial (the fourth of Lord Taylor's five essentials in Cowan). The appeal was allowed, since after the consideration of new evidence this court did not consider the Crown's case called for an answer. However, this court, presided over by Lord Bingham CJ, referred, in the language of the report's extract, to the model directions in these terms:

"The Court was reluctant to countenance the view that direction of a jury called for the mouthing of a number of mandatory formulae, and departure by the trial judge from a prescribed form of words would by no means always justify the upsetting of a jury's verdict. However, standard directions were devised to serve the ends of justice and the Court must be astute to ensure that these ends were not jeopardised by failure to give directions where they were called for. The drawing of inferences from silence was a particularly sensitive area…"

50. In R v. Chenia [2002] EWCA Crim 2345 (1 November 2002) on the other hand, where a section 34 direction again failed to contain the fourth essential relating to a case to answer, Clarke LJ said this (at para 53):

"It would in our view have been desirable for the judge to include a direction of the kind suggested, namely that no question of drawing an adverse inference could arise unless the jury were sure that there was a case to answer. However, we do not think that this trial could possibly be held to be unfair or the conviction unsafe on that ground. That is because no jury could possibly have concluded that there was no case to answer on the facts."

51. In the light of the current model JSB direction, it might be said that, in addition to or else in amplification or clarification of the statutory conditions emphasised in Argent and the five essentials emphasised in Cowan and Condron, the following matters should be set before a jury in a well-crafted and careful direction:

(i) The facts which the accused failed to mention but which are relied on in his defence should be identified: see para 2 of the model direction and Chenia at paras 87/89, where Clarke LJ said that this requirement must be approached in a common-sense way.

(ii) The inferences (or conclusions, as they are called in the direction) which it is suggested might be drawn from failure to mention such facts should be identified, to the extent that they may go beyond the standard inference of late fabrication: see para 2 of the model direction.

(iii) The jury should be told that, if an inference is drawn, they should not convict "wholly or mainly on the strength of it": see para 2 of the model direction and Murray v. United Kingdom 22 EHRR 29 at 60, para 47. The first of those alternatives ("wholly") is a clear way of putting the need for the prosecution to be able to prove a case to answer, otherwise than by means of any inference drawn. The second alternative ("or mainly") buttresses that need.

(iv) The jury should be told that an inference should be drawn "only if you think it is a fair and proper conclusion": para 3 of the model direction. This is not stated in the statute, but is perhaps inherent in that part of it emphasised in Lord Bingham's sixth condition. In R v. McGarry [1999] 1 Cr App R 377 at 383G this court glossed that condition as requiring a jury "not arbitrarily to draw adverse inferences".

(v) An inference should be drawn "only if…the only sensible explanation for his failure" is that he had no answer or none that would stand up to scrutiny: para 3 of the model direction, reflecting Lord Taylor's fifth essential in Cowan. In other words the inference canvassed should only be drawn if there is no other sensible explanation for the failure. That is analogous to the essence of a direction on lies.

(vi) An inference should only be drawn if, apart from the defendant's failure to mention facts later relied on in his defence, the prosecution case is "so strong that it clearly calls for an answer by him": para 3 of the model direction. This is a striking way to put the need, reflected in Lord Taylor's third and fourth essentials in Cowan, for a case to answer. A note, note 16, to the JSB guideline explains that it reflects "a cautious approach".

(vii) The jury should be reminded of the evidence on the basis of which the jury are invited not to draw any conclusion from the defendant's silence: see para 4 of the model direction and R v. Gill [2001] 1 Cr App R 11 at paras 30/31. This goes with point (iv) above, because it is only after a jury has considered the defendant's explanation for his failure that they can conclude that there is no other sensible explanation for it.

(viii) A special direction should be given where the explanation for silence of which evidence has been given is that the defendant was advised by his solicitor to remain silent: see para 5 of the model direction.

52. We now turn to the five defects of which complaint is made by Petkar, and mutatis mutandis by Farquhar (see para 16 above). We immediately observe that no complaint is made about Lord Taylor's first two essentials, or Lord Bingham's fifth and sixth statutory conditions, or points (i), (iv), (vi), or (viii) which we have derived from the model direction. Complaints are, however, made about Lord Taylor's third, fourth and fifth essentials, and about points (ii), (iii), (v) and (vii) above.
53. The first complaint, which reflects point (ii) above, is that the judge failed clearly to identify the inferences which the jury might draw. In this connection Mr Roche and Mr Clarke (the latter, for efficiency, essentially adopted the submissions of the former) relied on R v. J.O. (CACD, 9 June 2000, unreported) where Tuckey LJ said (at paras 20/22):

"20. Once the preconditions to the operation of section 34 are satisfied the jury are entitled to draw:

"such inferences from the failure as appear proper."

"21. The usual inference which the jury are invited to draw is that at the time of the interview the defendant had no answer to the allegations being made against him or none that would stand up to questioning. In other words, his subsequent defence is a late fabrication or one which has been tailored to fit the prosecution case. But the prosecution in this case did not invite the jury to draw such inferences. In his directions to the jury the judge does not say what inference the jury could properly draw if they decided to do so. We think he should have done. He should have reminded the jury of the inference which the prosecution invited them to draw along the lines of the specimen direction.

"22…It was incumbent on the judge in his summing-up to identify the relevant inference."

54. J.O. was a case where the accused of 16 had given a no comment interview on the advice of his solicitor. At trial the accused waived privilege and his solicitor gave evidence that the accused had given him before interview a very similar account of the incident to that which the defendant had himself given the jury at trial. The judge failed to tell the jury that they should only draw an inference against him if it was something which he could reasonably have been expected to mention. That was the critical defect found in the direction, but there was also the failure to identify the relevant inference referred to in the citation above. In the circumstances, it is not perhaps surprising that this court there found the conviction to be unsafe.
55. In the present case, however, we do not think that there is force in this first complaint. The judge did illustrate as possible inferences that "he had not thought it up, he had not decided on his defence, he had not decided whether to make use of that particular thing". However, he also said that he did "not wish to be specific here" and that ultimately any inference "is entirely a matter for you". Possibly deliberately, he did not cite expressly the standard inferences which are found in both paragraphs 2 and 3 of the model direction, viz that "he had no answer at the time or none that would stand up to scrutiny", which derive from within Lord Taylor's fifth essential in Cowan. On reflection, we think that these inferences, which are evolved historically from the section 35 direction, are more suited to the case of a no comment interview than to the interviews given by either of the appellants here (although they would, we suppose, match Farquhar's complete failure to suggest a defence based on duress). It can also be said that these standard inferences, perhaps for the very reason that they reflect a situation where an accused has effectively said nothing at all at his interview, if drawn, are likely to be particularly damning. In this case, on the other hand, the examples of inference cited by the judge well matched the significance of the appellants' individual interviews seen against their ultimate cases at trial. The inferences suggested by the judge are of the kind which might contribute to a jury being satisfied of guilt, but are not in themselves so critical. Although the judge would have been entitled to have pointed out that Farquhar's failure to mention a case of duress at his interview, if viewed by the jury as something which he could reasonably have been expected to mention, could well have invited the inference that "he had no answer at the time or none that would stand up to scrutiny", there is always the problem in cases involving co-defendants of moulding a direction which does not expose one defendant more sharply than the other.
56. An example of a case in which the inference canvassed was not the standard "no answer then or none that would stand up to scrutiny" is R v. El-Delbi [2003] EWCA Crim 1767 (20 June 2003), where (at paras 79/81) this court accepted the inference of "had not had a chance to prepare his story" as being its equivalent.
57. We conclude that although it was unsatisfactory for the judge to leave other possible inferences in the air, this complaint, if it stood by itself, would not cause us to regard the direction as a whole as defective. However, it is closely connected with the fourth complaint, to which we come below.
58. The second complaint is that the judge failed to tell the jury that they should not convict wholly or mainly on the strength of an inference. This reflects Lord Taylor's third essential, and point (iii) derived from the model direction. Mr Roche submitted that the judge's language that "the prosecution must still prove the case against them and they have got to have something by way of evidence, other than merely…that the defendant did not mention part of his defence" was unsatisfactory and insufficient. If it is so, it is because of the absence of that "or mainly". On balance we conclude that the language "must still prove the case against them" goes well beyond merely telling the jury that there must otherwise be a case to answer, and therefore that the judge's direction in this respect meets the spirit of what is required.
59. The third complaint is that the judge had failed to direct the jury that they could only draw an adverse inference if they were satisfied that the prosecution had made out a prima facie case (see Lord Taylor's fourth essential from Cowan). We consider that this requirement was well taken care of by that part of the judge's direction which has been quoted in the previous paragraph. We reject this third complaint.
60. The fourth complaint is that the judge failed to direct the jury that they should only draw an adverse inference if they concluded that the only sensible explanation for the defendant's failure to mention something was that he had no answer at the time or none that would stand up to scrutiny. Mr Roche referred to Lord Taylor's fifth essential, to para 3 of the model direction (see point (v) above) and to R v. Daly [2001] EWCA Crim 2643, [2002] 2 Cr App R 14 at para 13, where Kennedy LJ said:

"Turning to the second ground of appeal, we accept that if the jury was to be permitted to draw an inference a careful direction was required, for two interrelated reasons. First, there were matters on which the defendant was entitled to rely to explain the silence at interview, despite the form of the caution administered to him. [Sc Secondly,] At that stage he had not seen the video film, he was therefore unaware of the full weight of the prosecution case against him and in the light of his solicitor's advice he may have been understandably reluctant to admit a lesser but still serious offence. That made it particularly important in this case for the judge to say to the jury that they should only be prepared to draw an adverse inference if satisfied that the only sensible explanation for the appellant's failure to give in interview the explanation which he gave in his defence statement and at trial was that at the time of the interview he had no answer to the charge, or none that would stand up to questioning and investigation. It was important that the jury not be left at liberty to draw an adverse inference notwithstanding that it might have been satisfied with the plausibility of the appellant's explanation for his silence (see Condron v. United Kingdom (2001) 31 EHRR 1 at paragraph 61 and R. v. Betts and Hall [2001] 2 Cr. App. R. 251, at paragraph 48)."

61. The essence of this complaint relates to that part of the direction which insists that no inference should be drawn (see under para 45 above as to the possible inferences) unless there is no sensible explanation for the relevant failure to mention facts subsequently relied on. In the present case there is nothing in the judge's direction which reflects this requirement. And yet the defendants did put forward explanations for their failures, even if it would be understandable if the judge did not consider that they amounted to much. This complaint moreover goes together with the fifth complaint, which is that the judge should have reminded the jury of those explanations, which he only did in a separate part of his summing up where he dealt with the defendants' evidence and without relating that material back to the earlier direction on the right to silence.
62. We think that there is substance in the combination of these complaints and that this amounts to a misdirection, even though this case was quite unlike Daly. There the accused made no comment at all in his interview, in circumstances where he was so advised by his solicitor and because his difficulty was that although he accepted presence and theft of some cards, he did not accept the theft of cash, nor any threat of violence.
63. The question therefore arises whether this misdirection undermines the safety of the convictions, but before we reach that question we will consider the other grounds of appeal.

The second ground: a warning against the evidence of a co-defendant.

64. The judge gave no direction at all to the jury about how they should regard, in the case of each defendant, the evidence of his co-defendant. He did, however, give a full direction about treating each of the defendants separately. He said:

"Now, you do obviously need to make separate decisions on each defendant and each count…

"As I have said, each defendant in effect blames the other and says the other is guilty. But that does not mean that if you decide to acquit one of them and then move on to consider the position of the other, you must find the other guilty.

"What you need to do really is to consider each defendant separately and consider the evidence for and against him and make your decision. Up to you which order you decide to take them in. You may find that they are both guilty or one or the other or you may not be sure about either.

"If you end up saying, "Well, on a particular £420,000 occasion I can't be sure that Mr Petkar knew that this money came from [IBJ]. On the other hand, we cannot be sure that what Mr Farquhar did was not done through fear": well, then, you do not have to find someone guilty. You must acquit them both because each is entitled to a separate decision as if they – almost as if they were in the dock alone."

65. At the time of this summing-up, the JSB model direction 26 headed "Defendant's Evidence – Effect on Other Defendants" stated that where a defendant gives evidence in his own defence which damages a co-defendant's case or tends to implicate a co-defendant in the commission of the offence for which he is being tried, the jury should be warned about that evidence in some such terms as the following:

"The defendant A has given evidence which [damaged B's case] [tended to show that the defendant B was involved in some way in the commission of the offence(s) which you are trying.] Examine that evidence with particular care for A, in saying what he did, may have been more concerned about protecting himself than about speaking the truth. Bear that in mind when deciding whether you can believe what A has told you about B."

66. Of course, where each of two co-defendants gives such evidence, it might on that basis be thought that the judge would have to mould his direction to give matching warnings in favour of B against the evidence of A, as above, and in favour of A against the evidence of B. However, note 2 to this model direction read as follows:

"2. The above direction should not be given where co-defendants give evidence against each other. See R v Burrows [2000] Crim LR 48."

67. Prior to the Criminal Justice and Public Order Act 1994, it had been held that where one defendant gave evidence incriminating his co-defendant, just as in cases where an accomplice gave evidence for the prosecution, a full corroboration warning was desirable: R v. Prater [1960] 2 QB 464. That authority, however was questioned in R v. Knowlden and Knowlden (1983) 77 Cr App R 94, where it was held that that was not a rule of law but ultimately in the discretion of the judge: and that "the customary clear warning to examine the evidence of each co-defendant with care because each has or may have an interest of his own to serve" would in most cases suffice to ensure that the jury regarded the evidence in question with proper and adequate caution: per Watkins LJ at 100. In R v. Cheema (1994) 98 Cr App R 195, where the authorities were fully reviewed, Lord Taylor CJ said (at 202) that –

"The rule of practice that some warning, but not necessarily a full corroboration warning, is required where a witness, e.g. a co-defendant, may have a purpose of his own to serve was reaffirmed in Knowlden…"

and the passage at 100 from the judgment of Watkins LJ was cited.

68. In Burrows, however, Cheema was distinguished. There two defendants were accused of possession of cocaine with intent to supply. The cocaine was inside a capsule within an Easter egg. Each defendant blamed the other for putting it there, and professed his own ignorance and innocence. Both were convicted and one appealed: his sole ground was that the judge had not warned the jury that particular care was required when one accused gave evidence against another. The brief report states:

"Held, dismissing the appeal, that the judge had faced a stark difficulty. Any warning would have had to apply to both defendants, and would have meant directing the jury to treat each defendant's evidence with caution, just because it inculpated the other. That might have led to a complaint that the jury had not been allowed to approach the case with open minds. It was impossible in this case to give the normal warning. He did, however, underline the extreme care with which the jury should approach the allegations against each defendant, and in the circumstances the court could see no lack of safety in the verdict."

69. A note on the case by Professor Birch analysed the judgment in Cheema and continued (at 49/50):

"For all of these reasons, then, the right course of action in such a case was considered to be to steer what Lord Taylor called a "middle course", giving a milder form of warning to the jury that the co-accused might have an axe to grind (or words to that effect). It does not appear that his Lordship would have exempted from this the case where the whole thrust of A's evidence is to [place] blame on B (quite the reverse) and it is far from clear that he would have been persuaded to change his mind, at the time, if the facts were such that B's evidence was equally damning of A. But that was before the law was changed: the "axe to grind" warning is no longer a "middle course". Nowadays a Makanjuola warning, which might well take this form, is the most that can be given where an accomplice gives evidence for the prosecution. Thus it could well be that the case of the cut-throat defences equally requires a degree of reading down. Lord Taylor was an enthusiast for the abolition of corroboration warnings (see Cheema at p.205) and might well have ruled differently had the 1994 Act been in force.

"Be that as it may, the solution of the present case is clearly right if common sense is the guide. Most juries would already be distinctly unimpressed by a tale of grown men eating chocolate eggs, discarding the toys and, as if by chance, discovering an immediate use for the linings. When the point is reached at which the two linings become confused I certainly found it hard to stifle a giggle, and, had I been a juror, there would have been no need for me to be warned that both men might be trying to save their own skins by landing the other in it. Where the risk is so obvious, but the accused still hopes for an acquittal, the warning could be said to be prejudicial in that it comes too close to telling the jury what to find. Better to say nothing."

70. However, that two views of this are possible is indicated by the comment on Burrows of Archbold at para 4-404n:

"The reason given was that such a direction might indicate that the judge has formed a view about the way in which the evidence of each defendant should be approached. However, as argued at Criminal Law Week 99/29/9, it is submitted that this reasoning is unconvincing. It also seems inconsistent with previous authority."

71. Since then there has come the recent authority of Jones and Jenkins. There two defendants were convicted of murder. On the prosecution case it was joint enterprise; Jones' case was that both had indeed attacked the victim, but had caused him only minor injuries and that the fatal injuries had been caused later, perhaps in a road accident; Jenkins' case was that he had been present but not participant in any way, but in his evidence he acknowledged that the victim's injuries were serious (at para 9). Auld LJ described these as "modified cut-throat defences" (para 3). One of the grounds of appeal raised by Jones (but not by Jenkins) was that the judge had failed to give the "conventional" warning of the danger of accepting possibly self-serving evidence of one defendant incriminating a co-defendant.
72. Auld LJ said as follows:

"37. Mr Harrington [counsel for the Crown] submitted that the approach in Burrows is to be preferred to that in Cheema in the circumstances of this case, because Cheema was not a direct cut-throat case, whereas Burrows was…

"38. Mr Harrington also submitted that the judge's general directions to the jury as to how they should approach the evidence in this case sufficed in the circumstances. He referred to: the judge's direction…as to the need for separate treatment of the cases for and against each defendant, to his general direction…as to the need to consider the credibility of each witness in the case and whether it is self seeking or given to protect or to reflect badly on one defendant rather than another; and to his direction…as to the need for the jury to take the same care in their consideration of the evidence of each of the defendants as they did in respect of any other witness in the case. Those three directions, submitted Mr Harrington, taken together, were sufficient for the purpose.

"39. Whether the defences are "mirror-image" cut-throat defences, the law, since R v. Prater…has been that some such warning should normally be considered and given. Burrows was a case in which, as Judge LJ, giving the judgment of the Court said, "the difficulty facing the trial judge was somewhat stark". Any warning he might have given applied equally to each of the two co-defendants, whose cut-throat defences were almost a mirror-image of each other. Each had given evidence casting all possible blame on the other. It may be, as Judge LJ said, that within the confines of that particular case, the trial judge could not warn the jury to approach the evidence of each defendant with care because he had an axe to grind, without indicating to the jury that he had formed an adverse view about the way in which it should be approached by the jury. Though, why that was so, even in the particular circumstances of that case, is not readily apparent to me.

"40. A judge, even in a case of a mirror-image cut-throat defences, in the separate interest of each defendant, should be able to tailor a warning about the evidence of each against the other in a way that would not indicate that he, the judge, had formed an adverse view as to the defence of one or other or both. Even though the cross allegations are inextricably bound up in the defences of each, it is for the judge, in a neutral way, to give the jury such assistance as he can in their evaluation of the credibility of the evidence of each defendant as it is of that of all the witnesses in the case, whether for the prosecution or the defence.

"41. We see no reason to depart from the approach of this Court in R v. Knowlden & Knowlden…and confirmed in Cheema, that a judge, in exercising his discretion as to what to say to the jury should at least warn them, where one defendant has given evidence adverse to another, to examine the evidence of each with care because each has or may have an interest of his own to serve. Cheema was, as Mr Aubrey has observed, a cut-throat defence.

"42. There was also, as Mr Aubrey commented in argument, a particular need for some such warning in this case, where Jenkins, unlike Jones, had refused to answer questions in interview and was therefore able, if he wished, to tailor his defence to the facts in evidence.

"43. In our view, the failure to give such a warning was a serious omission and unfairly prejudicial to Jones' defence, and also, though possibly to a lesser extent, to that of Jenkins. Accordingly, we do not consider that the general directions as to evidence of the judge to which Mr Harrington referred us were sufficient for the purpose.

"44. Our attention has been drawn to current guidance of the Judicial Studies Board in the form of a note to its specimen direction No 26, which advises a form of warning to a jury where one defendant has given evidence which may have an adverse effect on a co-defendant. The guidance in the note is that such warning should not be given where co-defendants have given evidence against each other. The authority given for that proposition is Burrows.

"45. It follows from what we have said that we consider that no such general principle can be extracted from the case of Burrows, where it is plain from Judge LJ's judgment that the Court was heavily influenced by the facts of that case.

"46. Mr Aubrey has ventured an approach, which may be appropriate in many or most cases where a trial judge has to consider what if any warning to give where co-defendants have given evidence against each other. It seems to us to accord broadly with the general observations we have made about the principles derived from Knowlden & Knowlden and Cheema, subject always of course to what justice demands on the particular facts of each case.

"47. Mr Aubrey suggested that a judge, when dealing with the case against and defence of each co-defendant, might consider four points to put to the jury – points that would not offend any sense of justice and certainly would not cast the judge in the light of one who has formed an adverse view against either or both co-defendants. First, the jury should consider the case for and against each defendant separately. Second, the jury should decide the case on all the evidence, including the evidence of each defendant's co-defendant. Third, when considering the evidence of the co-defendants, the jury should bear in mind that he or she may have an interest to serve or, as it is often put, an axe to grind. Fourth, the jury should assess the evidence of the co-defendants in the same way as that of the evidence of any other witness in the case. That seems to us to be a useful – and suitably focused – approach when judges are faced with this particular problem, and we commend it."

73. Mr Roche and Mr Clarke naturally relied strongly upon all of these remarks. Mr Milne sought to distinguish them, on the basis that the present case was covered by Burrows. He accepted that in this case, unlike Burrows, the judge had not even warned the jury to approach the evidence of each defendant against the other with extreme care: but he nevertheless submitted that the full warning suggested by this court in Jones and Jenkins (as to which it was only the third constituent that was missing in the present case) involved the jury in mental gymnastics with a consequent danger that the value of each defendant's evidence in his own defence was watered down: this was particularly important to Farquhar, for only he could give evidence as to duress in his own defence. He submitted that the judge in this case faced the dilemma of the judge in Burrows and had given an adequate direction; and that in any event, although not stated explicitly, the fact that each defendant had an axe to grind must have been obvious to the jury.
74. We consider that there is much force in these submissions from Mr Milne, but also that it would be undesirable for us to depart, even if it were open to us to do so, from the decision of this court in Jones and Jenkins, and even though Auld LJ did qualify his remarks by reference to what justice demands on the particular facts of each case. We would, however, venture our concerns in this way. First, we would regard the danger of a warning regarding the evidence of cut-throat co-defendants not so much that which was canvassed and dismissed in Jones and Jenkins, namely that it may indicate to the jury that the judge had formed an adverse view as to their defences, but rather that it serves to devalue the evidence of both co-defendants in the eyes of the jury. It might be said that if the jury should regard the evidence of each defendant with a somewhat jaundiced eye on his "interest to serve" or his grinding axe, that goes far, in a real cut-throat defence, to undermine the defence of each.
75. Secondly, we wonder whether Auld LJ's third constituent to his direction (the warning) lies easily with his fourth constituent: and thus whether the distinction which a warning to be wary of a co-defendant's evidence is really designed to elucidate is that between evidence in a co-defendant's own defence (which has to be treated like that of any other witness) and evidence which incriminates the co-defendant. Of course, and this was the problem faced in Burrows, sometimes it is impossible to distinguish between the two. In this case, for instance, how could Farquhar give evidence in his own defence concerning duress inflicted on him by Petkar and his associates without thereby directly incriminating his co-defendant? And how could Petkar give evidence in his defence, namely that of innocent receipt of moneys to invest, without saying that Farquhar had told him that the moneys were his own to dispose of and thereby directly incriminating Farquhar? In other cases, however, the incriminating evidence may not be directly connected with a defendant's defence, and, interestingly, the example that can be given comes from Jones and Jenkins itself. The evidence of Jenkins that he had not participated in the assaults on the victim did not incriminate Jones at all: it was not as if Jones was saying that he was not a participant and it was all down to Jenkins. Jones' defence was rather that whatever he (and Jenkins) had done was not serious and not the cause of death. What incriminated Jones was Jenkins' admission, presumably in cross-examination, that the victim had received serious injuries. But that was not a necessary or direct part of his defence at all; nor is it at all clear why that admission may have been seen as serving his interests or grinding his axe. In these circumstances it might have been said that a warning in favour of Jones against his co-defendant's evidence might have been unfair to Jenkins. Considerations such as these suggest that the moulding of such warnings to the particular facts of each case are very difficult.
76. In sum, however, we consider there was a misdirection in failing to give any warning of any kind as to how to regard the evidence of each defendant in as much as it incriminated the other, albeit by no means as serious a misdirection as it might have been if this case had not been of a real cut-throat variety, in its own way not unlike Burrows.

Ground three: Mrs Wells' statement

77. This ground concerns only Farquhar. Mrs Wells had died between the first trial and the retrial. Her statement was therefore admitted into evidence under section 23 of the 1988 Act. What she there had to say was directed in support of Farquhar's defence of duress. She told of a number of phone calls from a man asking after her grandson: on one occasion the man asked if he was in Ireland and then said: "I am coming round to see you. I think it is time we had a chat." She replied that she would call the police. She was frightened and changed the locks. These phone calls ended in July 1999. She also told of an incident when she saw a man, whom some 11 months later in court on 13 June 2001 she claimed to recognise as Petkar, sitting in a parked car about 70 yards away. She told her grandson about these things after his return to England.
78. Farquhar's complaint is that the judge introduced her evidence in his summing up with the words that it was "not accepted", when that was true only on the part of Petkar, while the prosecution was merely neutral. It is submitted that the judge should have made this clear, especially as the burden of disproving the defence of duress lay on the prosecution. The judge said:

"No question of being unwilling to come to court but again this statement is not accepted and has not been tested in court."

79. In our judgment, however, the judge was concerned to mark the difference between the statements read under section 23 (such as Mrs Wells' and Ms Sweeney's), where there had been an issue as to their admissibility argued under sections 25/26, and other statements which were in truth "accepted" as dealing with matters which were not in dispute and which were therefore read under section 9 of the Criminal Justice Act 1967. The judge had made clear in each case as the various statements were read their difference in status, and in his summing-up he briefly did the same. Thus when he got to Ms Sweeney's statement, he said:

"You heard a statement read out from her. This one and one other [ie Mrs Wells'], which I will come to later, of course [are] different from all of the agreed statements that were read to you because this one of Karen Sweeney is not agreed so you have to consider what she says in it, bearing in mind it has not been tested in cross-examination in court…"

80. We think that in this context the judge's description of Mrs Wells' statement as "not accepted" is sufficiently accurate. It was not agreed. And even if the prosecution's attitude to Mrs Wells' evidence in itself was neutral, in as much as that evidence was relied on to support a case of duress the prosecution was certainly not neutral. We therefore reject this ground of appeal.

Ground four: the new evidence relating to Petkar's RBS account at Ealing.

81. Petkar's RBS account at Ealing showed the following. It was opened on 2 November 1999, with a deposit of £200. That was 11 months after the second transfer. The account then remained dormant until October 2000, when a further £50 deposit was made. It was now nearly two years after the transfers from IBJ. Then between October 2000 and 11 June 2002 there are 21 transfers into the account totalling some £177,000. Mr Clarke submits that these £177,000 may represent the greater part of the £222,000 which Petkar said went to Farquhar and which cannot otherwise be accounted for.
82. The 21 transfers fall into three main divisions. First, there were seven deposits between 19 October 2000 and 6 March 2001 of sums between a maximum of £7,500 and a minimum of £745.05 and totalling some £26,000, admittedly from unknown sources. Secondly, there then followed two much larger transfers of £30,000 each, one on 8 March 2001 and the other on 3 April 2001. These came from Petkar's Magenta Forex accounts. There were two such accounts. One was opened on 25 March 1999, the other on 7 March 2001. These accounts had large sums paid into them by named individuals or companies: Mr Clarke does not rely on any of those deposits as being from an impugned source. Between 1 September 2000 and 7 March 2001, that is to say during the period of the first seven transfers into the Ealing account, the first Magenta Forex account was dormant, the other had not yet been opened. It would seem therefore that those seven transfers totalling some £26,000 perhaps came from some other account being operated by Petkar, which has not come to light. Thirdly, there were a further twelve deposits between July 2001 and June 2002 of sums between £20,000 and £509. These are in the main traceable to Petkar's personal account at the National Irish Bank ("NIB"), which was opened on 8 June 2001 with a transfer of IR£359,000 from the second Magenta Forex account. In one or two instances the paperwork to render the NIB account as the source is missing, rendering the attribution less than certain, but the pattern suggests that these were also sourced from NIB and thus ultimately from the Magenta Forex accounts.
83. The court received a full file of accounting documents to explain these matters, together with a witness statement from DS Douglas Reeman, who conducted the police investigation into Petkar's financial affairs for the purpose of the confiscation order proceedings. The court, and Mr Clarke, also had the opportunity of questioning DS Reeman.
84. The overall effect of this material, therefore, is that the most that can be said is that transfers of some £26,000 only are from an unexplained source, and even those transfers occur some two years or so after the IBJ transfers. The Ealing account itself remained dormant for almost two years after the IBJ transfers. The court is therefore satisfied that the new material relied on by Farquhar, when seen in the context of all the new material which has come forward as a result of the confiscation order proceedings against Petkar, does not afford any ground for allowing the appeal. We do not think that this material, if it had been available to the jury, would have affected their view of the case in the slightest.

Safety of the convictions

85. We return therefore to the question of the safety of the appellants' convictions, in the light of our findings that there have been misdirections under the first and second grounds of appeal discussed above.
86. Mr Roche and Mr Clarke pressed on us their submission that in the light of those misdirections the court ought to find the convictions to be unsafe. They were able to point to a number of authorities to illustrate that submission, for instance Gill at para 32 and Daly at para 13, and Jones and Jenkins itself.
87. In our judgment, however, the misdirections we have found do not affect the safety of these convictions (nor, we might say, although no separate argument has been raised by reference to article 6 of the European Convention of Human Rights, do they render the trial unfair): see R v. Francom [2001] 1 Cr App Rep 17 at para 50 and Chenia at paras 58/59. The evidence against the appellants was overwhelming and we are satisfied that on the whole of the facts and with correct directions of law the only reasonable and proper verdicts would have been ones of guilty.
88. Thus in the case of Petkar, he accepted that he was the recipient of £840,000 (twice £420,000, before the first transfer was returned). He must have known that these monies came from IBJ and could not have come from Farquhar's own funds. Farquhar was only 28 and a relatively junior employee, who had not been working long at IBJ. The letter and statement from Barclays Bank told him that the money came from IBJ. On the basis that he was honest, he must have taken an interest in where so large a sum of money had come from, especially as he was concerned with the return of the first transfer. His explanation for the receipt of the money was that it was with him to invest for Farquhar for at least a year: nevertheless he immediately makes arrangements, in the case of both transfers, for spending large sums on himself or on repayment of his own debts, or for passing large sums through the hands of casinos. That in itself was wholly dishonest. Unaccountably, Farquhar requests the return of large sums, amounting in all to £222,000: those sums are returned in cash in such a way as to leave no trace. No receipts are obtained from Farquhar. The whole transaction, for not far short of half a million pounds, is done in the most unbusinesslike of ways, and without any documentation – other than may be found in the contract, whose terms are never carried out, and about which Petkar admits lying in his interview. In addition to that lie, there was the lie about having returned all the money to Farquhar, repeated to the Halifax, IBJ and the police.
89. Although this whole appeal has been premised on the basis that those lies amounted to failures to mention facts, they were in truth not so much failures to mention facts as lies about matters which were discussed. As such they were covered by a Lucas direction which has not been the subject matter of this appeal and which told the jury in terms that a lie was "only relevant if you are sure it was told because he realised he was guilty" and that "People may tell lies for reasons which have nothing to do with the offence". The jury were reminded about the explanations given by Petkar for those lies, even if that was done as part of the facts rather than directions of law. In these circumstances we view the judge's misdirection in connection with section 34 as being less serious than it might otherwise have been.
90. As for Farquhar, he admitted transferring the money and his only defence was duress. There is no complaint on this appeal about the directions regarding that defence. The evidence against him was equally overwhelming. The transcripts are wholly inconsistent with his acting under duress. Although he ascribed his departure to Ireland as an attempt to avoid Petkar, he did not so explain his use of a false name to open a bank account and buy a house or the change of his name to Ferguson. Moreover, Ms Sweeney's evidence was also very damaging, although we bear in mind that she had refused to come to England to give evidence and was not available for cross-examination. He also had to admit many lies: his pretence that he knew nothing about the transfers, for instance, or that he barely knew Petkar. Such lies, apart from their significance in themselves as redolent of guilt, were inexplicable if his true defence was duress. One then comes to his failure to mention the subject of duress at his interview: that is perhaps the one true example of a failure to mention a fact relied on by way of defence, but it is in its context a damning one. He had been away in Ireland for nearly two and a half years when he chose to return. He did so voluntarily and in his own time, as he says, even if his return was provoked by learning of the television programme which mentioned him. It is he, through his solicitor, who contacts the police. He is an intelligent man. It is inconceivable, in the absence of evidence of some telling explanation, that after all that time, and with the assistance of legal advice, he did not mention duress at the outset of his interview and well before the raising of the recorded and transcribed telephone conversations brought his interview to an end on his solicitor's advice. What then was his explanation for not then mentioning his defence of duress, even in outline? In effect there is no explanation at all: at most it is a pretence that he did not even know why the police had an arrest warrant for him, as he had heard, and that "I said as little as possible in the interview. I was pretty noncommittal". This barely counts as an explanation, and goes very far to undermine the defect in the judge's direction.
91. As for a Jones and Jenkins warning, even on the basis that one was needed, we think that on the facts of this case the absence of one was venial. As Professor Birch said in her note on Burrows: "there would have been no need for me to be warned that both men might be trying to save their own skins by landing the other in it". It is inconceivable that the jury did not understand the situation in the same way. In any event, as they had been told, the real decisions they had to take, subject to the burden of proof, rested on the credibility of each defendant's own evidence: in the case of Petkar whether he might have thought that Farquhar was an extremely wealthy young man who had really agreed to invest his own money with him for a year and then almost immediately asked for a little over half of it back; and in the case of Farquhar whether he might have been forced to steal from his employer against his will.
92. For these reasons we dismissed both appeals.

Petkar's appeal against sentence

93. We therefore reached the renewal of Petkar's application for leave to appeal against sentence. This was entirely based on the delay to Petkar's trial which had arisen out of Farquhar's absconding to Ireland. Farquhar had been hiding there for nearly two and half years, from December 1998 to April 2001. There was then a further delay, while a trial which was finally going ahead against Petkar alone was put back to accommodate Farquhar's return from Ireland. In the event the trial was itself put back a further nine months for reasons connected with an alleged defence of duress on the part of Farquhar which failed. Mr Roche therefore submitted that there had been a delay of nearly three and a half years through no fault of Petkar's own.
94. In R v. Clark [1998] 2 CAR 137 at 140E/F long delay "say, of over two years" between being confronted with dishonesty and the start of trial was cited as a matter of mitigation personal to an offender. We thought that there was sufficient in this point to merit a discount of 6 months in the case of Petkar on the five year sentence handed down to Petkar and Farquhar alike by the judge. We therefore gave leave to appeal and allowed the appeal to that extent.