The ethics and practice of witness preparations
Every day, in solicitors’ offices and barristers’ chambers around the country, witness evidence is being unwittingly corrupted. This is occurring because, acting within ethical boundaries and in accordance with general practice, lawyers are working to ‘refresh’ the recollection of witnesses and prepare them to give evidence by carefully walking the witness through the contemporaneous documents, asking leading questions, applying interrogative pressure and exposing them, directly and indirectly, to the recollections of other witnesses.
What is becoming increasingly clear to judges in the UK1 and Australia2, however, is that these common witness preparation practices can and do operate to distort and corrupt, not just the presentation of the witness’s evidence, but the witness’s actual recollection of relevant events, and do so in ways which are subtle, unconscious and not transparent. That these practices can corrupt the witness’s memory of the events is supported, not just by experience, but also by a substantial body of psychological research in which, over recent decades:
Hundreds of scientific studies have highlighted the fragile nature of witness memory and the ease with which memories can become unwittingly corrupted.3
Reflecting this concern, a working group of judges and practitioners was formed in the UK in 2018, to address the ‘fairly widespread feeling’ among users of the commercial courts that the written evidence-in-chief of witnesses was not the ‘best evidence’.4 The working group report of December 20195, concluded that the process by which witness statements were commonly prepared ‘may corrupt memory’ and recommended that this be addressed through an authoritative ‘statement of best practice’ for the preparation of witness statements. This recommendation ultimately resulted in the coming into force, from April 2021, of a new practice direction, 57AC.6
Practice Direction 57AC is an attempt to compel lawyers in the UK to change the practices commonly used to prepare witness statements in ways calculated to minimise the corruption of witness evidence. Since April 2021, witness statements for use in the Business and Property Courts of England and Wales7 have thus been required to comply with mandatory requirements8 that:
• witness statements must ‘be prepared in such a way as to avoid so far as possible any practice that might alter or influence the recollection of the witness’;
• interviewers must ‘avoid leading questions where practicable’ and ‘not use leading questions in relation to important contentious matters’;
• a witness statement must list all documents to which the witness has been referred for the purpose of providing the statement; and
• on ‘important disputed matters of fact’, the statement must ‘if practicable’ state in the witness’s own words ‘how well they recall the matters addressed’ and ‘whether, and if so how and when, the witness’s recollection in relation to those matters has been refreshed by reference to documents, identifying those documents’.
Practitioners and courts in the UK are still working through the full implications of these changes and it is not yet clear what impact they will have in practice, although it would appear that many lawyers are finding old habits hard to break.9 It seems doubtful, however, that this reform will effectively address the acknowledged capacity of the witness preparation process to corrupt witness recollection.
In order to appreciate why this is so, it is necessary to reflect on why the current practices are so widespread and entrenched. The duty of a litigation lawyer in the adversarial system (acting within ethical boundaries) is to obtain and put forward the best evidence possible to help win the client’s case. The lawyer is ethically bound not to ‘coach’ the witness, but is duty bound to question and test the version of evidence to be given by a prospective witness, including by drawing the witness’s attention to inconsistencies or other difficulties with the evidence.10 In practice, the line between ‘coaching’ and ‘testing and questioning’ can be elusive. In any event, the psychological research makes it clear that ‘questioning and testing’ a witness’s evidence in interview is more than capable of operating to distort the witness’s recollection, without anything approaching explicit ‘coaching’ having occurred.
In contrast to the imperative for strong and persuasive evidence, what the lawyer interviewing a witness commonly finds, particularly in a commercial case, is that the witness’s memory of critical events (e.g., representations made during the negotiation of an agreement, months or even years earlier) will be vague and fragmentary, at best. Typically, therefore, it is only by ‘refreshing’ the recollection of the witness, through extensive recourse to documents and other ‘prompts’, that any sort of coherent account can be obtained. By the end of that process, the witness’s evidence is, all too often, not so much a record of the witness’s actual memory, as a ‘careful reconstruction… based on a meticulous examination of all the documents in the case by the large teams of lawyers involved’.11 The evidence will, however, coherently ‘tell the story’ and will stand unless it is successfully impeached in cross-examination. If it is sufficiently credible and damaging, it may induce the other party to settle the proceedings, in which case it will never be tested at a hearing. The reality is that the efficiency of this process in turning the perceived ‘dross’ of vague, incoherent and fragmented memory into the ‘gold’ of coherent, consistent and complete evidence, means that it will not lightly be set aside by practitioners. No doubt it was a recognition of this that led the UK working group to recommend a mandatory statement of best practice, rather than expecting practitioners to embrace change voluntarily.12
The problem with setting strict rules, however, is that the witness preparation process is complex and nuanced, which makes it ill-adapted to a set of prescriptive rules governing all cases. This is reflected in the fact that most of the rules prescribed by Practice Direction 57AC are significantly qualified. Practitioners are thus, for instance, enjoined to avoid ‘so far as possible’ any practice which might alter or influence the recollection of the witness, use ‘particular caution’ in showing a witness a document they did not create or see at the time, use as few drafts ‘as practicable’ and generally avoid leading questions ‘where practicable’. Such qualifications, however, inevitably weaken the prescriptiveness of the practice direction and leave practitioners with considerable scope to continue with current witness preparation practices.13
In the few instances, however, where the requirements of the new rules are not qualified, this can also work against the objectives of the practice direction. Consider, for example, the new rule in the UK prohibiting interviewers from asking a witness leading questions in relation to ‘important contentious matters’. The rationale for this rests on an analogy between the interview process and oral evidence-in-chief, which must generally be given without the use of leading questions.14 However, the analogy between questioning to adduce oral evidence in court and interviewing a prospective witness is a poor one. In practice, the witness interview functions, not as a mere surrogate for oral evidence-in-chief, but as an integral part of the process by which a party’s lawyers investigate the facts and make decisions about what evidence to adduce. In that process, the lawyer does not merely reduce the witness’s evidence-in-chief to writing, but tests, probes and evaluates the witness’s version of events, in order to make judgments about whether or not to deploy their evidence. The reality is that this cannot effectively occur while avoiding asking the witness a question which is in any way ‘leading’.15 As a practical matter, this reality puts pressure on practitioners to find ways to ‘work around’ the prescription; e.g., by showing witnesses documents, or ‘indirectly’ pressing the witness on a contentious issue, or ‘clarifying’ their reaction to propositions in the pleadings or in the statements of other witnesses.
An alternative to specific and prescriptive rules, of course, could be a regime which identified the defects with current practice and required practitioners to exercise reasonable care to avoid those defects.
This approach is also seen in the practice direction, which requires those involved in the preparation of witness statements ‘to avoid so far as possible any practice that might alter or influence the recollection of the witness’.16 Diligently applied, however, this requirement would impose a formidable, unfair and probably impossible burden on practitioners.17 Among the practices identified by the psychological research ‘that might alter or influence the recollection of the witness’, for instance, are such common and reflexive features of witness interviews as using qualifying descriptors in questions (e.g., ‘how emphatic was he?’), interrupting a witness’s answers, applying interrogative pressure, giving implicit feedback to a witness on their answers (e.g., looking surprised or concerned), summarising the witness’s answer, or using documents to ‘fill in’ the witness’s chronology of events.18 Importantly, the science also demonstrates that a witness’s recollection of events can be altered or influenced by much more subtle cues than one might intuitively expect. In a seminal study, for instance, subjects were shown a video of a traffic accident and then asked to estimate the speed at which the car was travelling when it ‘smashed into’ the other car. This question yielded an estimated speed approximately 25 per cent higher than when subjects were asked to estimate the speed at which the car was travelling when it ‘contacted’ the other car (and, strikingly, also produced recollections of broken glass on the road when there was none).19
The practical reality is that, for a litigation lawyer to be fully informed as to the practices ‘that might alter or influence the recollection of the witness’, they would need to be educated and trained in the science of memory far beyond anything in existing legal education. Further, to have any chance of avoiding those practices, they would need extensive and rigorous training in interview technique.20 Faced with the impracticality of such a course, it seems likely that many practitioners will simply take refuge in applying a liberal view of the qualification that such practices must be avoided ‘as far as possible’.
A potentially more promising approach to the problem, which gains some endorsement in the practice direction, would be to give courts greater insight into how the witness was prepared to give their evidence, so that a judge can take that into account when assessing the likely reliability of the witness’s evidence. Practice Direction 57AC does this, in a limited way, by requiring that witness statements include a list of all the documents to which the witness has been referred in the interview process. On important disputed matters of fact, the rules also go further, to require the witness, ‘if practicable’, to state whether, and if so how and when, their recollection has been refreshed by reference to documents21, identifying those documents. The problem with this limited approach, however, is that it does not go far enough to hold out much prospect of making a real difference. This is because the insight provided into the witness preparation process by a mere list of the documents is likely in most cases to be too limited to be really useful and may even mislead. Without further insight into the way in which the interview process unfolded, it is difficult to see how a list of documents alone will much assist the court in assessing the reliability of the witness’s evidence. Where a witness’s evidence is contentious and may have been unconsciously corrupted by suggestion, in order to properly assess that possibility, one would need to know, for example, not just what documents the witness was shown, but all of the information that was shared with them, what they were told about the significance of particular facts to the case, what they were told about the recollections of other witnesses, whether they were repeatedly pressed to try and remember particular aspects of what occurred and how their evidence ‘evolved’ in successive interviews and drafts of their statement. Such an inquiry, however, is presently firmly forestalled by litigation privilege, which operates to ensure that communications with witnesses, notes of interviews and drafts of witness statements are all privileged.22
The promulgation of Practice Direction 57AC is an important, authoritative recognition that the process by which a witness statement is prepared can distort or corrupt that witness’s recollection. This is a positive development. Unfortunately, however, for the reasons set out above, the practice direction is unlikely to effectively redress this problem. As I have previously suggested23, there is a need to consider more radical reform; e.g., limiting the operation of litigation privilege over discussions with non-party witnesses. As French J observed 30 years ago:
The privilege attaching to statements taken from potential witnesses may not be supportable by public interest considerations of the same order as those enunciated in Grant v Downs24
… The confidentiality which attends their taking is of a limited character
… It may be that the time has come to reconsider whether such privilege as attaches to witness statements ought to continue …25
A strong argument can surely be made that, for both ethical and practical reasons, a party should not be able to engage in practices that have the real potential to corrupt the evidence of a witness and, at the same time, shield that process from substantive inquiry.26 This is an area which is ripe for reform. BN
1 See Gestmin SGPS S.A. v Credit Suisse (UK) Limited  EWHC 3560, - (Leggatt J).
2 E.g., J J Spigelman, ‘Truth and the Law’, Bar News, Winter 2011, 99, 109; Watson v Foxman (1995) 49 NSWLR 315, 318 (McLelland J).
3Kimberley A Wade & Ula Cartwright-Finch, ‘The Science of Witness Memory: Implications for Practice and Procedure in International Arbitration’ (2022) 39 Journal of International Arbitration 1.
4 https://www.judiciary.uk/wp-content/uploads/2019/12/Witness- statement-working-group-Final-Report-.pdf.
6 https://www.justice.gov.uk/courts/procedure-rules/civil/rules/part-57a- business-and-property-courts/practice-direction-57ac-trial-witness- statements-in-the-business-and-property-courts.
7 A collective of specialist civil courts which includes the courts of the Chancery Division, as well as the Admiralty, Commercial and Technology & Construction Courts.
8 Although mostly expressed in terms of what practitioners ‘should’ do, the legal representative responsible for the witness statement must certify that it has been prepared in accordance with the rules.
9 See, for instance, Mansion Place Limited v Fox Industrial Services Limited  EWHC 2747 (O’Farrell J); Blue Manchester Limited v BUG-ALU Technic GMBH  EWHC 3095 (Davies J); Prime London Holdings 11 Limited v Thurloe Lodge Limited  EWHC 79 (Thompsell J); Greencastle MM LLP v Payne  EWHC 438; Lifestyle Equities CV v Royal County of Berkshire Polo Club Ltd  EWHC 1244 (Mellor J).
10 Australian Bar Association, Barristers’ Conduct Rules, 1 February 2010, Rules 68 & 69; New South Wales Professional Conduct and Practice Rules 2013, Rules 24.1 & 24.2.
11 Estera Trust (Jersey) Limited & Anor v Singh & Ors  EWHC 1715 (Ch),  (Fancourt J).
12 Contrast the approach taken by the ICC Task Force in its November 2020 report on The Accuracy of Fact Witness Memory in International Arbitration, which merely recommended voluntary changes in witness preparation practice.
13 See J M Chin et al, ‘The New Psychology of Expert Witness Procedure’ (2020) 42 Sydney Law Review 69.
14 Statement of Best Practice, Appendix to PD 57AC, [2.5].
15 Noting that a ‘leading question’ is broadly defined as ‘a question that expressly or by implication suggests a desired answer or puts words into the mouth, or information into the mind, of a witness’ – Statement of Best Practice, Appendix to PD 57AC, [1.2].
16 Statement of Best Practice, Appendix to PD 57AC, [3.2].
17 Bearing in mind that the legal practitioner responsible for the statement must certify that it has been prepared in accordance with the statement of best practice.
18 See Kimberley A Wade & Ula Cartwright-Finch, ‘The Science of Witness Memory: Implications for Practice and Procedure in International Arbitration’ (2022) 39 Journal of International Arbitration 1. 19 E F Loftus, ‘Reconstructing Memory: The Incredible Eyewitness’, Psychology Today, December 1974.
20 Even experienced and trained interviewers can find this very difficult to achieve. A 1995 British study of actual police interviews, for instance, concluded that approximately 1 in every 6 questions posed by police interviewers to eyewitnesses was suggestive in some way – R P Fisher, ‘Interviewing victims and witnesses of crime’ (1995) 1 Psychology, Public Policy and Law 732, 740.
21 A requirement which it is suggested is likely in practice to result in a relatively formulaic approach.
22 See s 119, Evidence Act 1995 (NSW); Public Transport Authority of Western Australia v Leighton Contractors Pty Ltd  WASCA 151.
23 ‘Report of the Witness Evidence Working Group in the UK’  (Winter) Bar News 16.
24 (1976) 135 CLR 674. As the court explained, the rationale of privilege over communications between lawyer and client was ‘that it promotes the public interest … by encouraging the client to make a full and frank disclosure of the relevant circumstances to the solicitor’. As French J pointed out, this rationale does not extend to the protection of communications with non-party witnesses.
25 J Corp Pty Ltd v Australian Builders Labourers Federated Union of Workers (1992) 110 ALR 510, 515; see also Re Barings Plc, Secretary of State for Trade and Industry v Baker  1 All ER 673, 681-2. 26 The author is engaged in research towards a PhD on this topic and would welcome being contacted by any member of the profession who has encountered these issues in practice – firstname.lastname@example.org