Seaton v R

This appeal relates to comment made by the Crown on the omission of the defendant to call his solicitor. The case was one in which the defendant was accused of recent fabrication of his evidence on the grounds that he had said something significantly different in the police station. On his behalf, Mr Carey-Hughes QC contends that the comment made amounted to an infringement of legal professional privilege. In part the question raised is confined to the particular facts of this case and the course which the defendant’s evidence took. Mr Carey-Hughes, however, raises a number of questions about the import of the decision of this court in R v Wilmot (1989) 89 Cr App R 341, which has appeared in the textbooks for many years, when considered together with other decisions upon privilege as applied to the defendant in a criminal trial.

The facts of the case do not need detailed rehearsal. The defendant was charged with murder. The Crown case was that he had arrived home from Germany on the evening of Wed 24 Sep 2008, had visited his friend the deceased at the latter's home in Luton, and there had murdered him by multiple stabbing with a kitchen knife, a carving fork and perhaps a machete or similar, no doubt as a result of some altercation or dispute. There was some evidence of a possible source of dispute over money borrowed by the defendant to buy a property in Turkey.

The defendant's case at trial was that he was not there when his friend was killed. He had visited, but had left his friend alive and well and, after catching a nap in his car a few streets away, had gone home to Woolhampton in Berkshire, where he lived and worked at a public house called the Angel.

This had not been the defendant’s initial account. It had always been known and accepted that he arrived home from Germany late that evening, landing at Stansted not long before midnight and then being met by some friends and driven by them to collect his own car from their home in Romford. However, first thing the next morning he had told a friend of his at work that he had come straight home from collecting his car, but had been very delayed by motorway accidents. His route from Romford to Woolhampton would not take him to, or anywhere near, Luton. He was also interviewed by the police as a witness the day after the body was found. He told them the same.

By the time of the trial it was indisputable that the defendant had indeed been to Luton. There was evidence of sightings of his car where it would not have been if he had gone straight home but where it would have been if he had been to Luton. There was telephone evidence to the same effect. Cellsite evidence had put one of the deceased’s mobile telephones in the near vicinity of the defendant’s place of work well after the deceased had been killed and the defendant in due course admitted that he had been in possession of it. And most of all, the defendant's freshly spilled blood had been found in the deceased's flat.

At trial the defendant's case was that he had indeed visited the deceased, but had left him alive and well. The Crown asserted that the change of account was forced on him by the finding of evidence which put him in the deceased's flat. He said no; he had lied because he had been delivering a package of drugs and did not want to have to admit it.

A good deal of the defendant's blood had also been found in his own car. The Crown's case was that it derived from injuries to his hands, which he must have sustained in the course of using one or more knives to kill the deceased. He, by contrast, said that he had cut his hands innocently on various occasions, but particularly that he had cut one hand before he visited the deceased. So it was that his account of sustaining the injury to his hand assumed significance at the trial.

The injuries to the defendant’s hand had been observed when he was arrested about three days after the murder. In police interviews he was asked about them. He had a solicitor with him throughout. He elected to answer none of the many detailed questions which were put to him by the police, and it was apparent on the face of the transcript and thus in evidence that this was on the advice of the solicitor. Although he answered no questions, he did present the police, through his solicitor, with two prepared statements, one in the third interview and one in the fourth. They had plainly been prepared at the police station during consultations which had taken place between interviews. Of all the topics covered in the interviews, the only one on which he volunteered information in these prepared statements was that of injuries to his hands.

Each of his two prepared statements began with the following signed declaration:

“The statement below is my statement. My solicitor has gone through my statement with me and I agree its contents. No one has told me what to say.”

In the first prepared statement he said that he had cut his hand at Stansted airport at baggage reclaim when, as he tried to remove his bag from the carousel, a woman passenger had accidentally dropped her own bag on his hand.

At trial he said this was wrong. It had been at Lubeck airport on departure that he had cut his hand, not at Stansted on arrival. And it followed that it had not been at the carousel, but rather when a fellow passenger had dropped a sharp piece of luggage onto his hand as he was going through security.

In his second prepared statement the defendant gave further information about other injuries to his hand seen at the time of his arrest. One, he said, had been sustained sweeping away broken glass in the pub after returning there, thus a day or so after the murder. This second statement came during the fourth interview, and it had clearly been prepared during a consultation with his solicitor of about half an hour between the third interview (at which the first statement had been delivered) and this one. Although this second prepared statement dealt also with injuries, it was silent as to the matter of airport injury; in particular it did not correct what the defendant now told the jury was an accidental misstatement in the first statement.

The Crown case was that this was a further example of the defendant altering his evidence to meet the indisputable facts. They said he had altered his account in the face of evidence that (1) there was a lot of blood in his own car which must have come from his cut hand but (2) none in the car in which his friends had taken him from Stansted back to Romford. Moreover those friends recalled no sign of any cut, nor reference to one. That meant, said the Crown, that the original assertion that the cut had been sustained just before getting into the friends’ car, and some little time before getting into his own, could not be correct. Counsel for the Crown had other examples of changes in the defendant’s account to explore, not least the earlier denial of going to Luton at all, and the jury would have to decide whether the defendant’s explanation might be truthful or not. But it is clear that counsel for the Crown took the stance, in due course, that this particular change in the defendant’s story was one for which there was no plausible explanation, and that it was, for that reason, a good touchstone as to his truthfulness.

The defendant’s evidence at trial on this topic proceeded as follows. First, he was asked about it by his own counsel in chief. He gave evidence that he had been advised not to answer the policemen’s questions, but did not adduce the terms or contents of that advice. He was asked directly whether the assertion in the first statement that the accident to his hand had occurred at Stansted was correct. He said it was not. The questioning proceeded:

“Q: Who wrote out whatever it was that was said?

A: My legal representative.

Q: We know that your signature appears on it, along with a declaration that you have read it and it is correct.

A: Yes

Q: Did you read it?

A: I didn’t, no.

Q: Why not?

A: Erm, I should’ve done – in hindsight. Erm, but to tell you the truth I was so tired, erm…I’d obviously been in the police station. I didn’t sleep. Erm, and with that and my medication made me even worse – I actually went to the hospital that Monday as well. So, I didn’t feel too well – too well.

Q: Anyway, you did sign it without reading it.

A: Yeah

Q: It was read out during the course of the interview

A: Yes

Q: When it was read out, did you appreciate that it said that you had injured your hand at Stansted Airport, picking up luggage off a carousel?

A: Yes

Q: Was that correct?

A: No, it wasn’t.

Q: Did you realise….

A: I did at the time

Q: …that it was wrong? And did think about correcting it?

A: I did, but then I thought, erm…

Q: I do not mean during the interview but at any time immediately after the interview

A: Yeah. After.

Q: I just want you to say ‘yes’ or ‘no’ to this, please: did you discuss the possibility of correcting it with your solicitor? Just yes or no please.

A: Yes

Q: In the light of whatever discussion you had with your solicitor did you come to a decision about whether you change it there and then, as it were?

A: Yeah, but I decided not to change it at the time.”

The defendant went on to say that he had reached that decision because he thought the information irrelevant, and hoped shortly to be released, but felt that if he changed the statement the police might not believe him. Inferentially at least it appears that he was saying that he thought it might delay his release to alter the statement.

In cross examination, counsel for the Crown established that this meant that the defendant was saying that the responsibility for the (as he now said) error in the first prepared statement was that of his solicitor, who had made a mistake. Counsel made it clear to the defendant that his conversations with his solicitor were privileged and that he could not and did not ask to see her notes. The reference to privilege troubled Mr Carey-Hughes, who raised with the judge, in the absence of the jury, the issue of how far cross examination could go. In the event, after discussion, it did not go beyond underlining the suggested improbability that the error should not be corrected when the obvious opportunity afforded by the second prepared statement presented itself immediately after the error had been noticed by the defendant. However, in discussion with counsel both then and later during the trial the judge ruled that counsel for the Crown was entitled to comment to the jury on the absence of the solicitor from the witness box. Counsel for the Crown did make that comment, with some force. He did so in the context of a speech in which the change of account in relation to the injury was treated as a good touchstone for the truthfulness of the defendant and for what was suggested to be the manner in which he had altered his account in several respects in order to tailor it to the indisputable evidence. It is that comment which Mr Carey-Hughes contends was impermissible, and which forms the basis of his appeal to this court.

This question arose unexpectedly. The only authority to which the judge was referred was Archbold and in particular to a reference to R v Wilmot. The report was not available immediately, though it may have been consulted as the trial proceeded. There has been debate before us as to what proposition Wilmot is properly authority for. As will be seen, it makes in the end no difference to the result of the present case which it is. Having had the benefit of argument, however, we think that we should set out our analysis of Wilmot because it appears that it may have been assumed without argument in subsequent cases to have said something which it is not at all clear that it did say.

Wilmot was charged with a series of six predatory rapes, committed by picking up women, some prostitutes, in one or other of two cars. On appeal, the principal issue was the cross admissibility of the evidence of the several complainants and whether it had been proper to try the counts together. That depended on the law of similar fact evidence as it stood before DPP v P [1991] 2 AC 447 and long before the bad character rules now contained in the Criminal Justice Act 2003. The appeal on that point failed. It is the second point which concerns us.

The second point related to the defendant’s change of account. When first arrested he had said he had had nothing to do with any of the complainants. When told of the possibility of scientific evidence, he confessed to the rapes. But in interview the following morning, attended by a solicitor, he reverted to his original denial of any connection with any of the complainants. He retracted the confession, saying that he made it to ensure the release from the police station of his girlfriend. At trial, however, his evidence was that he had had sexual intercourse with each of the complainants, but that it had been consensual, followed by a dispute. Whatever else he had previously said, he had never said this. The Crown’s case was that this was a recent fabrication, designed to accommodate evidence which clearly put him in the company of the complainants – there was fingerprint evidence, semen analysis and the finding of property belonging to one of them which made a strong case that he had been present.

Counsel for the Crown had cross examined the defendant by suggesting that this was recent fabrication. There was nothing arguably wrong with that. But he had also asked the defendant: “Did you tell your solicitor the truth?” The defendant declined to answer the question. In due course counsel for the Crown commented on the absence of the solicitor from the witness box and the judge told the jury this, dealing with the allegation of recent fabrication:

“When that sort of suggestion is made, the law is this. Until such a suggestion is made, nothing that an accused has said privately to his solicitor or his counsel or his Mum or his Dad about the case is admissible. But the minute it is suggested that it is a cook-up, a fairly fresh invention, an accused is entitled to give and to call evidence in proof of the fact that he had indeed said what he is now saying to you at a much earlier date.

……That suggestion…entitles an accused both to give and call evidence that the suggestion of recent invention is all rubbish, that privately and at a much earlier date he had said it to his solicitor, and his solicitor is entitled to come in to the witness box and say the same thing. You have heard no such evidence and therefore you are entitled to ask the question “Are we sure [that he is speaking the truth]?”

It is to be noticed that the question asked by counsel for the Crown came at a time when the defendant had said nothing whatever about his solicitor, or any discussions with him, or indeed anything at all about when his present account of matters had first seen the light of day. Counsel’s question went directly to a privileged occasion. There had been nothing which could arguably have constituted a waiver by the defendant of his privilege, and when asked the question, the defendant had declined to answer.

The court dismissed the appeal. One view of this case is that it is authority for the proposition that no question either of privilege or waiver arose in the circumstances of that case. The relevant part of the headnote in the Criminal Appeal Reports certainly begins:

“(2)

The case did not raise a question of breach of privilege at all. The privilege prevented the prosecution from calling the solicitor to give evidence but the appellant was entitled to call him, just like any other witness, to rebut the allegation of recent fabrication.”

The second sentence is obviously correct. The appellant was entitled to call his solicitor, just as he could call any other witness. The first sentence makes a broader statement altogether, namely that no breach of privilege arose anywhere in the case. It is certainly possible to see how this has been extracted from the judgment, but it is by no means clear that it is what the court meant to say.

The court set out the submissions of Crown counsel which were in effect twofold:

Next, the court went on to hold that there had been a very strong case of recent fabrication, and that to suggest it was completely justified. It added “We emphasise that in our view it is not a question of breach of privilege at all.” That might mean that there was no breach of privilege anywhere in the case, or it might mean that if the defendant had called his solicitor that would not have amounted to a breach of privilege.

Since the court went on to point out that the Crown could not call the solicitor, and only the defence could have done so, that is some indication that it meant no more than that there would have been no breach of privilege had the defendant called him. More significantly, the court went on specifically, and emphatically, to hold that the question asked by Crown counsel in cross examination (“Did you tell your solicitor the truth”) ought never to be asked. The court expressly said that such a question put the defendant into “an impossible position”. The only basis on which that question was wrong was that it intruded upon legal professional privilege at a time when the defendant had himself done nothing which could even arguably amount to waiver of his privilege. The reason why it put the defendant in an impossible position was, plainly, that it wrongly forced him either to waive privilege or to suffer the criticism that he must be untruthful because he had not. In other words, the court was recognising that privilege indeed existed, and that Crown counsel’s question had intruded onto it.

The true reason for the decision of the court, on this secondary ground of appeal, appears from what follows its unequivocal rejection of the propriety of counsel’s question. It held that in the context of an overwhelming case against the defendant, the asking of one improper question “cannot be thought to have made an appreciable difference of any kind to the conclusion which the jury reached which was amply justified by the evidence.”

It appears that it may be that in some subsequent cases, Wilmot (perhaps on the basis of its headnote) has been taken (without detailed analysis) to mean that a defendant faced with an allegation of recent fabrication can call his solicitor without that constituting any element of waiver of his privilege. Before considering those cases, it needs to be recorded that nowhere in the judgment in Wilmot is there any discussion whatever of waiver. That is unsurprising, since Wilmot had not done anything approaching waiver: see paragraph 19 above; what he had done by refusing to answer his cross-examiner’s question was not to waive, but rather to claim, privilege. At most, the court spoke of whether there had been breach of privilege, and as to that, as we have seen, the better view is that it was saying no more than that the defendant himself could not be in breach of his own privilege.

For the reasons which we have given, it seems to us that the court in Wilmot was not holding that no privilege existed at all; plainly it did. It cannot have been expressing the view that counsel’s question did not intrude on privilege, for if it did not, it could not have been roundly criticised, as it was. If the court was expressing the view that no breach of privilege would have been involved if the defendant had opted to call his solicitor on this point, that was obviously correct, for the privilege is that of the defendant and he can rely on it or not at his election; he cannot be in breach of it.

The four subsequent cases which we need to examine were all concerned with section 34 Criminal Justice and Public Order Act 1994, which of course did not exist at the time of Wilmot.

R v Condron and Condron [1997] 1 Cr App R 185 was the first case in which this court gave general guidance on the approach to the new section 34. The defendants were charged with the supply of heroin. They had declined to answer police questions and it was on the record that their solicitor had advised them not to do so, on the grounds that he considered them unfit because they were displaying withdrawal symptoms; the doctor who examined them had disagreed. The principal decision of this court was that where it was in evidence that the defendant had been given legal advice not to answer questions that did not mean that no adverse inference under section 34 could ever be drawn, but that a direction based upon the one appropriate to failure to give evidence (as explained in R v Cowan [1996] 1 Cr App R 1) ought to be given. However, on the facts of the case, although the direction had not been as complete as the court now explained it should have been, the convictions were entirely safe.

The court went on to give general guidance on the conduct of cases where section 34 arose, dealing with matters which are now uncontroversial, such as the extent to which the Crown should adduce the fact of a ‘no-comment’ interview, the scope of cross examination upon failure to mention matters now relied upon, and the questions which should be posed to the jury. It also made (at 197A-G) some general observations upon legal professional privilege.

Those observations contained the following propositions:

It will be seen that the last proposition (v) contains two statements. The first, that a defendant is free to call anyone he likes to prove that he gave the disputed explanation at some previous time, is wholly uncontroversial, and it is this which is said to be based on Wilmot. It is also, as we have shown, what Wilmot said. The second statement, in the final sentence, is an addition. It looks as if it was thought to be derived from Wilmot, because of the statement that the solicitor is in the same position as anyone else, which may well derive from the submission of Crown counsel in Wilmot. But in fact Wilmot had said nothing at all about waiver, as we have shown.

Propositions (ii) and (iii) from Condron were re-affirmed in R v Bowden [1999] 2 Cr App R 176. The defendant was charged with robbing a McDonald’s restaurant. He had refused to answer questions when interviewed on arrest, and his solicitor had put on record that this was on the grounds that the solicitor did not think the evidence strong enough. At the trial, the defendant adduced the terms of that advice. The questions which he had declined to answer included enquiries into an apparent sudden increase in wealth, without obvious source, shortly after the robbery, and about his having taken a holiday immediately afterwards in Gran Canaria where he had been photographed in celebratory pose outside the local branch of McDonalds. At his trial he gave detailed explanations both for his spending and for wishing to photograph the McDonalds restaurant. Accordingly the question arose whether an adverse section 34 inference was open to the jury or not.

This court confronted directly the issue whether section 34 had altered the law of legal professional privilege. It held that it had not. It re-affirmed the fundamental nature of the privilege, relying inter alia on the well known decision in R v Derby Magistrates ex p B [1996] 1 Cr App R 385, then comparatively recently decided. The court confirmed the distinction drawn in Condron between adducing merely the fact of legal advice to remain silent and adducing the content or terms (i.e. the reason) for that advice. The first does not waive privilege; the second does.

The decision in the case was accordingly that this was a case of the second sort. Privilege had been waived. It was open to the Crown to question the defendant about what he had told his solicitor and, when he said he could not remember what he had told him about the money and the photograph, that could be taken into account by the jury in deciding whether or not the real reason for not giving the police the explanation now advanced was that it was a recent invention, rather than the reason given by the solicitor for the advice.

The difficulty arises in relation to the following general observation at 182C, which followed endorsement of the general rule of privilege, but was not integral to the issue or the decision in the case:

“It is well established that the privilege is that of the client. It is waivable, but only by or on behalf of the client. A waiver ordinarily occurs when a client chooses, for whatever reason, to reveal the effect of a communication protected by the privilege; he cannot claim privilege for that which he has voluntarily revealed. It makes no difference whether the revelation is made by the client or by the legal adviser acting within the scope of his authority as agent on behalf of the client. Nor does it matter when the disclosure is made.

When a defendant at trial deposes to facts which he has not mentioned at an earlier stage and it is suggested to him that these facts are an invention or fabrication after the event, the defendant may rebut that accusation by asserting and calling evidence to show that he mentioned the facts to another person at that earlier stage, and no waiver of privilege is involved even if evidence is given (by him or his legal adviser) that this disclosure was made to his legal adviser. Such an accusation was made against the defendant in Wilmot….”

The emphasis is ours. The emphasised words were clearly said to be based on Wilmot. The judgment appears to have treated the court in Wilmot as having approved the whole of Crown counsel’s submission (summarised at paragraph 22 above). The court went on to derive the same proposition from Condron, which it treated as having in turn derived it from Wilmot.

In R v Wishart [2005] EWCA Crim 1337 the Crown submitted that the distinction made in Condron and Bowden between adducing the fact of legal advice to remain silent and adducing the contents of the advice was wrong, and that there was a waiver of privilege implied in either type of case. That contention was plainly barred on the authority of both Condron and Bowden, the latter of which was directly binding on this court on the point, and it was duly rejected. In that case, the Crown had gone a great deal further than in any other reported case. The defendant had relied on an alibi not advanced in police interviews. He had put in evidence the fact that he had had legal advice not to answer questions but had not adduced any evidence of the contents of or reasons for that advice, so that this was a case of the first type identified in Condron and Bowden. Accordingly he had done nothing to waive privilege. Nevertheless, in cross examination counsel for the Crown had asked exactly the same question as had been roundly disapproved in Wilmot, viz whether the defendant had told any of his legal advisers what he was now saying. No-one told the defendant that he did not have to answer that question. The defendant had said that he thought he had told his solicitor. The judge had then directed the solicitor to disclose the whole of his records of the consultation at the police station (i.e. upon any topic whatever). Unsurprisingly, that procedure was held to be wrong.

In referring to Wilmot, Condron and Bowden this court cited the passages from the latter two cases which we have set out above at paragraphs 31(v) and 36. It made this comment:

“16.

It could be argued that in such circumstances what a defendant says to his solicitor is a privileged communication and that he waives privilege in that communication by volunteering that he has made it. But that is not what this court has said in the three cases to which we have referred which have not been doubted in any more recent cases where the earlier decisions have been followed. Even if we thought that there was some reason to doubt this well-established line of authority, we are bound to follow it.”

In fact, as we have seen, it made no difference to the decision in Wishart what the correct analysis is of a defendant who voluntarily opens up the topic of what he told his solicitor, since Wishart did not voluntarily do anything of the kind.

Loizou [2006] EWCA Crim 1719 was an example of the second type of Condron/Bowden case, where the defendant had put in evidence not only the fact of legal advice to remain silent but also the reasons for it. On appeal the issue was whether her having done so opened the door to the question in cross examination: ‘Did you tell your solicitor any of the account that you have given today?’ The answer was that on the facts of that case it did. The basis for that was that once the defendant had opened up the suggested reason for her silence, it allowed opposing counsel to probe whether that reason could possibly explain not saying (as she now was saying at trial) that she was an innocent dupe who had no connection with the transaction in question except as an interpreter.

On the basis of Wishart, the court assumed, without questioning it, that to call one’s solicitor to rebut recent fabrication involved no waiver of privilege (paragraph 65), but that issue did not arise in the case. The importance of the decision lies in the scholarly analysis which begins at paragraph 66 and which demonstrates that waiver, when it occurs, is not all or nothing. Waiver can be, and often is, partial. In every case of waiver, the question what the waiver has let in is determined by the test of fairness, or, to put it another way, what is necessary to avoid there being left a misleading impression by revelation of part only of the privileged communications, or the defendant, in colloquial terms, ‘having his cake and eating it’. The test adumbrated in Auburn on Legal Professional Privilege (2000) at page 216 was approved; namely that the essential question is whether the partial disclosure has actually led to unfairness or prejudice. Thus, for example, in Burnell v BTC [1956] 1 QB 187 a cross examiner who put to a witness part of a previous statement he had made to his solicitor could not resist disclosing the whole of the statement. And as Hobhouse J held in General Accident v Tanter [1984] 1 WLR 100,

“if evidence is adduced then the extent of the waiver relates to the transaction to which the evidence goes….It does not extend to all the matters relating to the subject matter of those conversations”

The court might equally have been referred to B v Auckland District Law Society [2003] 2 AC 736, where Lord Millett at paragraph 68 said this:

“It does not follow that privilege is waived generally because a privileged document has been disclosed for a limited purpose only……It must often be in the interests of the administration of justice that a partial or limited waiver of privilege should be made by a party who would not contemplate anything which might cause privilege to be lost and it would be most undesirable if the law could not accommodate it.”

It should be noted that this is not the exercise of the s 78 Police and Criminal Evidence Act jurisdiction, although the result may well be the same. This needs to be understood, because s 78 applies only to evidence which the Crown seeks to adduce, whereas the question what has been let in by a waiver of privilege may equally arise on the case of a co-defendant.

For the foregoing reasons our view is that Wilmot is authority for the proposition that a defendant who volunteers evidence of his communications with his solicitor does not breach privilege; the privilege is his to waive. It is not authority for those communications not being privileged. Nor is it authority on the question of when a volunteering defendant waives the privilege. For the same reasons, although later cases, and particularly Bowden, assumed that Wilmot had said that voluntary giving of such evidence in order to rebut an allegation of recent fabrication entailed no waiver, this was unnecessary to any of the decisions and proceeded upon an untested assumption born of the juxtaposition of different propositions in earlier judgments, particularly in Condron. .

The foregoing analysis leads us to the following conclusions:

Mr Carey-Hughes contended that all communications between the defendant and his solicitor were privileged, that there had been no waiver, and that accordingly the comment which the Crown made was impermissible. He submitted that if and insofar as Wilmot decided that there was no question of privilege and that for that reason the comment could not amount to a breach of it, the case was wrongly decided.

As we have endeavoured to show, Wilmot did not, properly analysed, decide that there is no question of privilege or of waiver when a defendant gives evidence that he told his solicitor the account which he is now producing at trial.

However, whatever the correct analysis of Wilmot, this is a clear case of waiver by the defendant’s evidence in chief. The defendant’s evidence that his first prepared statement contained an error, and that he had signed it without reading it, can mean nothing other than that he had told his solicitor something different and the responsibility for the error was hers. That was putting in evidence his suggestion that he had told her, from the outset, not Stansted but Lubeck. Even more clearly, his evidence that he had noticed the error when the first prepared statement was read out, and had discussed with her correcting it, could mean nothing other than that he had told her, between the two interviews, ‘not Stansted but Lubeck’. We do not agree with Mr Carey-Hughes that it is possible to avoid this obvious conclusion by the ‘forensic device’ of asking no more than whether the defendant had discussed the error with his solicitor and not explicitly what he had said. There was no point in asking the defendant whether he had discussed the “error” with his solicitor unless it was to inform the jury that the allegation of recent fabrication was unfounded because the defendant had always said what he was now saying at trial – ie Lubeck. He could not do that without waiving privilege. It is clear that when the argument arose unexpectedly at the trial, and without any of the authorities being available apart from Archbold’s reference to Wilmot, the point was not argued on the basis of waiver. But that makes no difference. It is not as if the point was argued before the defendant gave evidence and he relied upon what was argued when he came to give it. By the time the argument arose, he had given his evidence. He had deliberately put before the jury the plainest possible suggestion that he had not changed his account but had given his solicitor the same one as he was giving the jury. That inevitably must be characterised in law as a waiver.

The Crown did not in fact probe any further in cross examination into the content of the communications between the defendant and his solicitor. The question before us relates solely to the comment on the fact that if the defendant’s evidence were true, it could and would be confirmed by the solicitor, who was available. Once there had been this waiver, this comment was, on the facts of this case, wholly justified. We agree that adverse comment on the absence of a witness is by no means always justified. Often the jury would be in real danger in deducing conclusions from the absence of a witness, because there may be many good reasons why he cannot or will not be called. Moreover in the case of many, but not all, witnesses, either side could call them. Those were no doubt among the reasons which lay behind the decision in R v Wheeler (1967) 52 Cr App R 28 that all that ought normally to be said is that neither side had called the witness. But as this court explained in R v Shakeel Khan [2001] EWCA Crim 486, each case depends upon its facts and it may be entirely fair to comment on a witness not being called where the evidence shows that he is available and could readily give the lie to a suggestion made against the defendant if it is false. It is true that, in strict theory at least, once there had been the waiver that there had, the Crown could perhaps have called the solicitor themselves, limited to the point in issue. But in reality that would not be practicable. The solicitor could not be proofed. Even if, technically, it were possible to by-pass the general rule requiring service in advance of the statement of a rebuttal witness (as to which we say nothing), there would be considerable danger, if she were called blind, of trespassing inadvertently beyond the part of the communications which the defendant had opened up. The reality in this case was that if the defendant’s evidence of what passed between him and his solicitor was true, it was the simplest possible thing for him to call her, and she would have laid immediately and conclusively to rest the suggestion of recent fabrication of the account of the cut to the hand. Her absence, on the facts of this case, was indeed telling, as the Crown submitted, and since the defendant had opened up the topic, the jury was entitled to consider her absence.

If, contrary to our view, the law is that the evidence the defendant gave did not impinge upon privilege at all, then the same result follows.

For these reasons we hold that the comment made by the Crown was not impermissible. If it had been, it was sufficiently significant in the context of this case to make the conviction unsafe. Since it was not, the conviction is perfectly safe and the appeal must be dismissed.