It may seem strange to link the highly important Court of Appeal decision in Warren & Ors v R. [2021] EWCA Crim 413 with modern civil litigation. However this is an important case and there are direct parallels to many issues we see in contemporary civil cases. The important point here is to retain draft statements. The drafts will be privileged but you never know when reference to the drafts will be necessary.

If the destruction of the handwritten statements had been revealed to the appellants at the time of the trial, this issue could have been comprehensively investigated with the witnesses when they gave evidence, and the judge would have been able to give appropriate directions. We have no doubt that if that had happened, the trial process would have ensured fairness to the accused. Self-evidently, that is not what occurred. Instead, we are confronted with a situation in which an unknown number of the first written accounts by eyewitnesses have been destroyed in a case in which the allegations essentially turned on the accuracy and credibility of their testimony.”


In 1973 and 1974 a number of individuals were convicted of conspiracy for acts that allegedly took place in the building strike in 1972.  The Court of Appeal allowed an appeal against these convictions. It was found that the prosecution were relying on witness statements that had been added to and amended over time. There had been significant additions to the statements.  The fact that the original statements were not available at the trial meant that the convictions were not safe.



58. The West Mercia Police Report (see [20] above) sets out many of the problems that faced the investigating police officers, which had led to prolonged enquiries with a view to tracing and prosecuting those responsible for the violence. The report rehearses in significant detail the areas covered in the various witness statements that had been taken, highlighting the parallel investigations by different teams and the cooperation between them. A schedule was prepared, which included, inter alia, a description of the events at the various sites, a list of the witnesses in each instance, together with a summary of their individual accounts together with any identifications that had been made. The police relied substantially on press photographs, including of the marches that had been held to promote publicity concerning the strike, instances of peaceful picketing and the events on 6 September 1972 at Telford. These photographs were shown to all the witnesses “with a view to identifying persons involved in disorderly picketing – both in Shropshire and North Wales“. In a section of the report, entitled The Statements, the following is set out:
“104.The statements taken from witnesses fall into four main categories (viz. non-striking workmen, miscellaneous witnesses, pickets and police officers). All these statements, whilst in Criminal Justice Act form and signed, include matter which may be held to be irrelevant and also some hearsay. By the very nature of the investigation, this form of statement was considered essential in the initial stages and is left on file for the information of counsel.
(a) Non-striking workmen
105. All workers on the seven affected building sites were interviewed and statements recorded, embodying evidence of the disorder and damage and, where applicable, the identification from the photographs of those responsible. These constitute the majority of witnesses.
106. One point must be made here: due to the circumstances, confusion, and fear generated by the pickets, it would be unrealistic to think that all these could have been identified. Indeed, in the circumstances we have been singularly fortunate in the number that have been identified.
(c) Pickets
110. […] it was decided to interview all the identified “passive” pickets.
111. Several […] made statements […]
112. One point to be made about these statements is that these men visited so many sites and […] finer details is short on accuracy. Basically, however, the story they tell is corroborated by other witnesses.
(d) Police Officers
116.These are largely non-evidential as to specific offences but do fill in the background of the sites and […] details of the interview with the accused persons.”
    1. The report goes on to suggest that some of what occurred was spontaneous and had not been planned in advance, for instance the events at Kingswood, Shelton roadworks, The Mount and Severn Meadows. Similarly, the visit to Telford was said to have been spontaneous. The authors then added:
“126.The evidence against several […] organisers and leaders […] is not so strong. It mainly consists of the very act of organising their party’s attendance, in circumstances where disorder on a large scale must have been foreseen, and the fact that they were present on the sites with the pickets without trying to restore order (or paying lip-service in that respect). […]”
    1. Mais J ordered that before entering the court to give evidence the witnesses should see their witness statements and the photographs that they had earlier viewed for identification purposes.
    1. As set out above, during the first trial, and this was likely to have been the general position in Trials 2 and 3, the appellants accepted their presence at the various picketing sites but challenged the identification evidence to the extent that it was suggested they had been involved in criminality. As Mais J summed up the defence contention:
“The accused, on the other hand, say that they took no part in any violence; they threatened no one; they acted peacefully; they did no damage. They say that if there was any violence, any threats, any intimidation, any damage, they were not parties to it, they did not lend their support and were not responsible in any way.”
    1. To a significant extent during Trial 1, and again in all probability repeated in Trials 2 and 3, the witnesses were cross-examined on the basis of their statements, and any contradictions and omissions were highlighted. Additionally, Mr Platts-Mills (appearing for Mr Warren) during his closing speech made significant criticisms as to what he suggested was the selective and partial approach of the police. The judge dealt with this issue during the summing up as follows:
“Mr Platts-Mills on behalf of Warren said this: there was a partial and selective weeding of witnesses. Admittedly the prosecution only called 200 witnesses. The police had interviewed and obtained statements from some 700. The Defence have been provided with the names of all such, so it is said. It is the duty of the Prosecution to adduce relevant evidence before you. That is the duty of the Prosecution. The Defence, providing they are given the facilities, know what other people have said. It is up to them, if need be, but the Prosecution’s duty is to produce the relevant evidence before you.” (our emphasis)
    1. Before the three trials, the police revisited some witnesses in order to obtain, inter alia, identification evidence from them. Unlike the modern practice of taking a further or additional statement, the police – potentially, we observe, in many instances – elected instead to amend the earlier statement. A witness called Roger Castle provides an example of this. The first version of his statement that has been produced is dated 29 September 1972. In it he described the actions of some of the pickets, including what was allegedly said by their “spokesman“. He gave no description of this individual. The statement was then amended on 13 March 1973 (as indicated in the body of the statement) when PC Jones visited Mr Castle and showed him a number of photographs. This section of the statement, as it is to be inferred, commences with the words “I have now been shown some photographs marked A – N”. As set out in what is apparently, therefore, an amendment to the statement, he suggested that one of the individuals was the spokesman, who had been dressed in a white shirt and (he thought) in jeans, and who had a Welsh accent (see [48] above).
    1. We interpolate to observe that this was a significant alteration, not only because it involved his identification of Mr Tomlinson as the spokesman, but he also described what he said was his clothing and accent.
    1. The same course was apparently taken with a witness called Henry James (one of those picketing: see above at [31]). His statement was purportedly taken on 2 November 1972, although the date does not appear on the version of the document provided to defence counsel. There is no indication in the body of the statement that it was altered, but in evidence Mr James said that a section had been added to the end of the statement without his consent. There was no subscription, as with Mr Castle’s statement, to indicate it had been altered. It is perhaps notable that the section that Mr James suggested had been added begins with the words “I have now been shown a set of photographs lettered A – N“.
    1. Again, the importance of this development is that there had been a significant change in the witness’s account during the process of seemingly adding to the original written/typed account, in that in the addendum Mr James asserted that Mr Pierce was one of the front runners at Telford and had charged up the site waving a stick. Mr James said that he had not seen this alleged occurrence and that it had been added to his statement without his permission.
    1. It is unsurprising, therefore, that during the first trial, the appellants were forensically interested in the existence of any additional accounts from the witnesses. In support of this appeal, they rely on four instances when the issue arose as to the availability of earlier statements or reports from particular witnesses, which resulted in a response from the judge or prosecution counsel that would have had a tendency to deter them from pursuing this line of enquiry. In essence, two things occurred. Either the prosecution strongly implied that the appellants were in possession of all the potentially relevant materials, or the judge indicated that if the issue was pursued the witness’s statement would be made an exhibit and provided to the jury (in accordance with the practice at the time), or alternatively he chastised defence counsel for pursuing the issue.
    1. By way of detail, during the cross examination of PC Jones, Mr Drake stated in front of the jury that all the statements, including those the defence were not entitled to see, had been made available. When Mr Turner-Samuels Q.C. on behalf of Mr Carpenter was asking Mr James about the addition, without his consent, to his witness statement as set out above (at [65]), the judge observed that this had involved “a most outrageous suggestion” by counsel, notwithstanding the fact that it was the witness who had volunteered this information. During the cross-examination of Alan Hordley, Mr Platts-Mills attempted to test the reliability of the witness by reference to changes in the account that he had provided to the police. The copy of the witness’s statement served on the defence in advance of the trial was dated 20 March 1973. However, on the last page there was a subscription: “Statement amended from statements taken on 7th and 13th September, 1972“. Mr Platts-Mills was concerned to highlight how the witness’s account had changed during this process, and particularly that in the statement of 7 September 1972 he had not mentioned the “the big man” (Mr Warren) “taking a poke at him“. This statement had not been provided to Mr Platts-Mills until Mr Hordley was in the witness box and was being questioned by him during the trial. Mr Drake made it clear that if it was to be suggested that the witness’s account had changed, the relevant statements should be provided to the jury. When Detective Inspector Gradwell gave evidence, he indicated that he had dictated a statement on 7 September 1972. However, the statement he was shown during the trial, dated 30 March 1973, was not his original statement. The statement of 7 September 1972 does not appear to have been available. We note that he looked at photographs to make identifications within a few days of 6 September 1972.
    1. It is on the basis of these clear foundations for the interest on the part of defence counsel in any changes in the accounts of the witnesses as to what had occurred, along with the apparent assertion by the prosecution that the accused had been shown all the statements – including those they were not entitled to see – that we turn to the lynchpin of this ground of appeal. This is to be found in paragraph 16 of a note of a consultation on 17 September 1973 at which Mr Drake and officers from West Mercia Police were present (held at Mr Drake’s home). This document was found in the National Archives in late October 2013. The note was prepared by an assistant chief constable (administration), Alex Rennie, who was present at the meeting. The note was sent with a covering letter dated 20 September 1973 to Mr Desmond Fennell, Mr Drake’s junior, who had not attended the consultation, as well as to the office of the Director of Public Prosecutions. Paragraph 16 sets out:
“So that Counsel would be aware it was mentioned that not all original hand written statements were still in existence, some having been destroyed after a fresh statement had been obtained. In most cases the first statement was taken before photographs were available for witnesses and before the Officers taking the statements knew what we were trying to prove.”
    1. It is important to understand the extent to which the terms of this note reveal what must have occurred. The incident was on 6 September 1972. The relevant photographs (from the press) were in the possession of the West Mercia Police by 13 September 1972. However, it is by no means clear by which date the officers understood what it was that the prosecution were trying to prove. Therefore, if the destroyed “original handwritten” statements in the first sentence, which represented “some” of the total, are the same as the “first statement(s)” in the second sentence (which seems to us to be the likely position), “most” of them would have been written at a time before the police had the press photographs (within 7 days of the incident) and before the police knew what those responsible for the prosecution were trying to prove. In our judgment it would be erroneous, therefore, to conclude that this subset of destroyed handwritten statements had necessarily been provided before 13 September 1972. The most that can be said with confidence is that there were handwritten statements from eyewitnesses which had been destroyed once i) the police were able to show the press photographs to the witnesses and ii) the officers knew what those responsible for these prosecutions were seeking to prove in the forthcoming trial.
    1. One of the investigating forces, Gwynedd police, adopted the straightforward approach, which mirrors that of today, in which they took initial statements to which were added later additional statements. The statements taken by West Mercia Police, by way of contrast, were in various forms. Some were taken on a particular date and they provide no indication as to whether later amendments or additions were made. Others are dated March or April 1973 but are recorded as having been taken in September or October 1972. Some were clearly marked as being further statements. In other instances, it is evident that composite statements had been prepared, which combined the contents of various statements that had been prepared on more than one occasion (for this latter category, see [63], [65] and [68] above). A number of statements had details added to them (for instance under the heading “further states“). In other instances, the handwritten original statements are still in existence, a number of which were exhibited at trial.
    1. Critically, however, none of the surviving statements indicates it is a replacement document, in the sense described in paragraph 16 of the note of 17 September 1973, namely that an earlier statement, or a version of it, had been destroyed leading to a replacement statement. We note that the committal proceedings were delayed because the statements “were still being amended and signed following Counsel’s advice“.
    1. The CCRC considered whether it is possible to establish the approximate number of statements to which the note refers but concluded this would be a futile exercise.
    1. There is an absence, therefore, of any reference in any of the documents in this case, save for paragraph 16 of the note of 17 September 1973, to this procedure of destroying an unknown number of the original hand-written statements. As we have already set out above, the report of the West Mercia Police to the Director of Public Prosecutions (“DPP”) dated 18 December 1972 summarises the difficulties faced by those responsible for the investigation and the reliance on photographs shown to witnesses when taking the statements. It records that “the task would been virtually impossible without Press photographs: of various marches held to promote publicity in the strike, of actual peaceful picketing, and one set taken on a Telford site on “Black Wednesday” by the local news photographer […]” (see paragraph 102). It is of note that the Chief Constable of West Mercia Constabulary made no mention of the destruction of some statements in his letter of 18 December 1972 to the DPP which attached the report. Instead, he set out not, we observe, entirely accurately:
The West Mercia Investigating Officers were fortunate in having press photographs of the pickets and they started by identifying persons on the photographs and then taking statements when identifications had been made. The officers considered that all statements taken should be included to present a full picture and also to enable counsel to discard those not required.”
    1. The Chief Constable added, “The officers are in no doubt that there may be some difficulties in identification after this lapse of time”.
    1. The prosecution, with considerable industry for which we are grateful, have sought to demonstrate that there was a clear and consistent pattern to the way in which the civilian eyewitness statements were taken. There were over a hundred and forty individuals in this category. Mr Price Q.C., leading counsel for the Crown, has set out a number of key factors which include:
i) There was clear utility in indicating in the margins of the witness statements the identity of the individuals identified in the photographs;
ii) Only the last witness statement from the witness was placed in the trial bundle (this is described by Mr Price as an annotated account of a witness, recorded in a single document);
iii) Statements sometimes incorporated the contents of earlier statements or referred to other statements (e.g. George Evans’s statement is dated 21 March 1973 but at the conclusion it is set out “Statement taken at […] at 8.35 on Thursday, 7th September, 1972; and amended […] at 5.00 p.m. Wednesday 13th September, 1972; Trevor Clarke’s statement is dated 3 March 1973 and at the conclusion it is recorded “Statement taken at 11.55 a.m. on Thursday 28th September, 1972 […]”;William Allen’s statement is dated 21 March 1973 but at the conclusion it is recorded “Statement taken from statements taken on the 9th and 20th September 1972“; Robert Briscoe’s statement is dated 20 March 1973 and it concludes “The details contained in this statement were originally reported to Police on 7th September 1972, and added to on 3rd October 1972“; George Stubbs’s statement is dated 21 March 1973 but at the conclusion it is recorded “Statement amended from statements taken on the 8th and 14th September 1972“).
    1. It is contended that it is reasonable to infer that the destruction of an original handwritten statement only occurred after the replacement had been taken and only when the latter contained the information set out in the original. It is suggested that:
“Even if a statement taken prior to 13.09.72 had not already been typed and or photocopied by the time it was destroyed, its content was otherwise preserved in its replacement. That it should by then not have been copied or typed, is in any event submitted to be highly unlikely, not least because it would have been expedient when going to see a witness for a second time, so soon after the first with a photograph album, that the officer should also take with him a copy of the first statement, with which to begin the process of supplementing the narrative with additional information obtained from a review of the photographs”.
    1. The case against all the appellants was essentially based on the testimony of eyewitnesses who were asked to look at photographs of potential suspects some days at least (but in some cases it might have been weeks or months) after the incidents had occurred. These events substantially predated our era of ubiquitous CCTV cameras and mobile telephones, which frequently provide a contemporaneous and continuous record of public events. Cross-examination in the circumstances of the present case, particularly in the absence of modern methods of verification, can be critical. One of the vital means of demonstrating that an eyewitness is unreliable is by careful examination of the opportunities the individual had for observation; their powers of perception and memory; mistakes they have made in recalling and recording what occurred; inconsistencies in their evidence; and omissions or inconsistencies revealed in, or by, previous statements. Proof of previous inconsistent statements is governed by sections 4 and 5 Criminal Procedure Act 1865. Although criminal cases are infinitely various, based always on their particular facts, comparing and contrasting the various accounts of eyewitnesses can have a substantial – indeed, potentially determinative – impact on their credibility.
    1. We are unpersuaded by Mr Price’s analysis of the witness statements and the conclusions that he suggests should be drawn. We can see no basis for concluding that the content of a destroyed witness statement would necessarily have been preserved in its replacement. Indeed, we would suggest that the opposite may – indeed, was likely – to have been the case, given the destroyed statements in all probability had a different focus than their later iterations, since they were taken before photographs were available and before the officers taking the statements knew what the Crown were seeking to prove. Furthermore, Mr Price’s contentions have been substantively undermined by the transcript of the cross-examination of Alan Hordley by Mr Platts-Mills. As set out above, Mr Platts-Mills sought to test the reliability of the witness by reference to changes in his account. The witness statement dated 20 March 1973 was an amendment from statements taken on 7 and 13 September 1972. Mr Platts-Mills was provided with the statement of 7 September 1972 during his cross-examination of the witness, and it was clear that the witness had not mentioned in this earlier version key assertions that featured in the later statements, for example that the “the big man” (Mr Warren) had taken “a poke at him“.
    1. Similarly, again as set out above, Mr James testified that there had been a significant change to his statement when additional detail was added to the original iteration. It was suggested that Mr James had claimed that Mr Pierce (a Trial 2 appellant, William Michael Pierce) was one of the front runners at Telford and had charged up the site waving a stick. Mr James said that he had not seen this alleged occurrence and that it had been added to the end of his statement without his permission. The layout of the statement potentially supported Mr James’s contentions in this regard, demonstrating that additional allegations may have been added, once the photographs were available and the officers understood what the prosecuting authorities were seeking to prove.
    1. In our view, these two examples exemplify the lack of a proper basis for the Crown to assert that we should infer that nothing of consequence was lost in the process of destroying this unknown number of original handwritten statements. The respondent’s detailed analysis set out above, albeit presented skilfully and helpfully by Mr Price, does not therefore support the conclusion that the contents of the destroyed statements would necessarily have been preserved in the later statements. Considered realistically, recollections on an unquantifiable number of occasions will have changed and additional details will have been provided as the statement-taking process unfolded, in a case which involved a large number of eyewitnesses.
  1. If the destruction of the handwritten statements had been revealed to the appellants at the time of the trial, this issue could have been comprehensively investigated with the witnesses when they gave evidence, and the judge would have been able to give appropriate directions. We have no doubt that if that had happened, the trial process would have ensured fairness to the accused. Self-evidently, that is not what occurred. Instead, we are confronted with a situation in which an unknown number of the first written accounts by eyewitnesses have been destroyed in a case in which the allegations essentially turned on the accuracy and credibility of their testimony. As we have already described, we consider it correct to infer that the descriptions by the witnesses would in many instances have changed and developed as they were shown the photographs and as the police gained greater understanding of what those responsible for the investigation sought to prove. Those changes and developments could have been critical for the assessment by the jury of whether they were sure that the individual appellants were guilty of the charges they faced. The jury either needed to have this evidence rehearsed in front of them to the extent necessary, if the statements were still in existence, or they needed to be given clear and precise directions as to how to approach the destruction of the statements if that had occurred. Neither of those things happened, and in consequence we consider the verdicts in all three trials are unsafe. The common law has developed significantly in this area over the last half century, particularly as regards the obligation on the prosecution to retain a record of any variations in the statements of relevant witnesses (see the Code under the Criminal Procedure and Investigations Act 1996 (“CPIA”), paragraphs 4 and 5) and to disclose them if that material might reasonably be considered capable of undermining the prosecution’s case or assisting the case for the accused. It is to be stressed that under paragraph 5 CPIA Code, the duty is to retain the final versions of witness statements and draft versions where the content differs, along with, inter alia, any material casting doubt on the reliability of a witness. As in Bentley, this court in arriving at this conclusion on the first ground of appeal has applied “legal rules and procedural criteria which were not […] applied at the time“. By the standards of today, what occurred was unfair to the extent that the verdicts cannot be upheld.


A lawyer never knows when someone is going to dispute the accuracy of a witness statement. It can be long after the event.  The new Practice Direction on Witness Statements in the Business and Property Courts gives a requirement that all drafts be kept.

5.10 Wherever practicable –

(1) a trial witness statement should be based upon a record or notes made by the relevant party’s legal representatives of evidence they obtained from the witness,

(2) any such record or notes should be made from, and if possible during, an interview or interviews (using any convenient format, for example face to face meeting, video or telephone call or conference, webchat or instant messaging),
If a trial witness statement is based upon evidence obtained from the witness by other means (for example by written answers to a questionnaire or the exchange of emails or other forms of correspondence, or by the witness preparing their own draft statement), the guidance set out in this Appendix should still be followed, so far as possible and modified as necessary.

5.11 An interview to obtain evidence from a witness –

(1) should avoid leading questions where practicable, and should not use leading questions in relation to important contentious matters,

(2) should use open questions as much as possible, generally limiting closed questions to requests for clarification of or additional detail about prior answers, and

(3) should be recorded as fully and accurately as possible, by contemporaneous note or other durable record, dated and retained by the legal representatives.