Legal development

Re-examining the Approach to Factual Witness Evidence in International Arbitration


    This article summarises a chapter first published in the Stockholm Arbitration Yearbook 2023. The publication is available at Kluwer Arbitration here (Chapter 5, pp. 75 – 88).


    Witnesses have always played a central role in the determination of legal disputes in all major legal systems. But the way in which witness evidence is assessed, its primacy (or lack thereof) and how it is deployed and verified differs significantly, not least between the civil and common law systems.

    Generally speaking, common law jurisdictions adopt an adversarial approach to witness evidence, placing a focus on what the witness says and that testimony being open to probing and challenge through cross-examination. In contrast, civil law jurisdictions tend to place greater emphasis on documentary evidence and the adoption of an inquisitorial approach where witnesses are subject to relatively limited questioning by a judge or other decision-maker instead of by an opposing party’s counsel.

    International arbitration, which benefits from an ever-broadening diversity of practitioners from different legal systems, has supported a ‘reconciliation’ of approaches through ‘soft’ laws such as the IBA Rules on Taking of Evidence in International Arbitration that provide parties with a flexible framework for management of witness evidence. However, in the authors' experience, common practice in international arbitration reflects predominantly the common law approach with frequent use of lengthy written witness statements often prepared with significant assistance of lawyers. The common law approach of using witness statements has, particularly in recent years, been subject to criticism in failing to provide the best evidence at proportionate cost.

    Purpose of witness statements and challenges in presenting witness evidence

    The Witness Evidence Working Group, set up under the auspices of the Commercial Court Users' Committee, was tasked with assessing factual witness evidence in trials before the Business and Property Courts in England and Wales. It found that only 6% of survey participants thought that the current system of witness statements ‘fully’ achieved the aim of producing the best evidence possible, with 45% of participants considering that it did so only partly or not at all (Report, p 9). The Witness Evidence Working Group outlines that a consistent theme emerges: the ‘over-lawyered nature of witness statements’ (Report, p 9).

    At the same time, it seems clear that the answer is not to dispense with written witness statements altogether. In fact, a survey conducted by the School of International Arbitration at Queen Mary University of London and White & Case LLP (2012) found that 92% of respondents opposed eliminating witness evidence (Survey, p 29).

    The answer, therefore, lies in changing the process by which witness statements are prepared in regard to the science of memory. The ICC Commission’s Report on the Accuracy of Fact Witness Memory in International Arbitration (the ‘ICC Report’) highlighted the following areas of concern in this regard.

    (a) The impact of phrasing on responses to questions

    Specific wording used in witness interviews can materially change the evidence that a witness recounts. However, standard practice in the preparation of witness statements involves a witness being questioned or interviewed by a lawyer who may prepare a first draft of a witness statement based on the answers given by the witness. Therefore, the manner in which a witness is questioned can potentially have a material impact on his or her evidence.

    (b) The misinformation effect

    Is ‘a phenomenon where typically misleading information which participants are exposed to after an event interferes with or impairs their original memory of that event’ (ICC Report, p 11). This means that evidence given by witnesses may not be accurate as a result of prior discussions between the witness and others involved in the dispute. At its extreme, the ‘misinformation effect’ has been shown to result in individuals ‘recalling’ entire fabricated events which happened to them personally.

    (c) The impact of retelling on subsequent recall

    Research has shown that taking a particular perspective after an event can later lead to a biased recollection of the event. in the context of a commercial dispute, where a witness is often engaged in telling his or her story from a particular perspective, there is a real risk of a biased retelling (ICC Report, p 13). 

    A suggested way forward

    The authors suggest that the following steps can be taken in appropriate cases to ensure that witness evidence is necessary, accurate, helpful, and prepared without unnecessary costs being incurred. (ICC Report, section V, p 20). 

    (a) Is witness evidence necessary?

    Parties and counsel should consider whether or not witness evidence is necessary. It is unnecessary for witness statements simply to repeat what is set out in contemporaneous documents.

    Generally, witness statements are required to: (i) prove disputed facts which cannot be proven by other documentary evidence; (ii) explain the context of certain documents, e.g., a telephone conversation which preceded an email exchange; (iii) provide background, e.g., explaining the efforts made by the founders of a company which led to goodwill in the market; and (iv) providing technical experience – although this would typically be the subject of expert evidence (ICC Report, section IV.B, p 17). 

    (b) Is accuracy important and is there a risk of contamination?

    Counsel should consider taking alternative approaches to drafting witness statements depending on the extent to which: (i) accuracy in a witness’ evidence is important; and (ii) there is a risk of contamination.

    Accuracy is likely to be important where there are disputed facts and limited contemporaneous documents, such as where negotiations have taken place over a telephone call. But if there is a low risk of contamination because there is only one individual with the relevant information, then it may not be proportionate or appropriate to take all of the steps set out below.

    (c) Limited direct evidence

    The practice of limited direct evidence from a witness combined with a witness statement gives some ‘colour’ to the written testimony and also means the witness is more ‘invested’ in the process of giving evidence and so less likely to delegate the exercise of preparing a witness statement to lawyers. Knowing that a witness will also be giving direct evidence (which can never be fully controlled by the lawyer in a way that written statements can) should serve to instil some caution and mean that written statements more closely align with the witness’ factual knowledge.

    (d) Witness summaries

    The Witness Evidence Working Group has suggested the use of witness summaries instead of full witness statements. Under this approach each witness would briefly outline the facts within his or her knowledge that are relevant to the issues in dispute, but would not go into extensive detail and would not refer to all of the documents. (Review of Civil Litigation Costs: Preliminary Report, p 407). 

    The authors do not consider that witness summaries should replace witness statements. Instead, the authors are of the view that it may be appropriate to consider whether a witness should be first asked to prepare a summary of his or her evidence, which then forms the basis of interviews with lawyers according to which a witness statement can be prepared. Such an approach would minimise the risk of inaccuracies in witness recollection and would allow for the preparation of a witness statement that seeks to clarify rather than alter or interfere with a witness’ evidence.

    (e) Witness interviews

    Witness interviews should be conducted in a manner that is similar to a direct examination. This would mean using open-ended and not leading questions to minimise the impact of phrasing on responses (ICC Report, section II.A), interviewing witnesses individually to minimise contamination of memory and keeping an accurate record of the interview so that any subsequent witness statement can accurately reflect the witness’ evidence (ICC Report, section V.A(ii)).

    (f) Preparation of witness statements

    In England, trial witness statements are now to be prepared in accordance with the Statement of Best Practice, in the Appendix of Practice Direction 57AC of the Civil Procedure Rules. This requires that ‘[t]he preparation of a trial witness statement should involve as few drafts as practicable’ (Appendix, para. 3.8) because ‘[a]ny process of repeatedly revisiting a draft statement may corrupt rather than improve recollection’ (Appendix, para. 3.8). Witness statements should also be as far as possible in the witnesses’ (Appendix, para. 3.6) own words and should be based on the record of the interviews with the witness and the witness summary (Appendix, paras 3.9 and 3.12).


    Civil and common law jurisdictions have over the centuries developed their own practices in relation to factual witness evidence. Both approaches have their advantages and disadvantages, but international arbitration enjoys the privileged position of being able to choose on an ‘á la carte’ basis the best elements of each.

    Practices such as greater use of a discerning approach taken to how witness statements are to be prepared and deployed and usefully drawing from the experience of reforms in national court systems should be encouraged.

    Arbitration practitioners – particularly tribunals – should increasingly challenge assumptions as to how witness evidence should be deployed efficiently and fairly, exploiting to the fullest degree the flexibility that is said to be one of international arbitration’s greatest strengths.


    Stay ahead with our business insights, updates and podcasts

    Sign-up to select your areas of interest