New ICC Report on The Accuracy of Fact Witness Memory in International Arbitration

Undoubtedly, the year 2020 will remain unforgettable for the whole world. Nevertheless, the particular memories of this period will obviously differ for every person that has experienced it. The human memory is based on a subjective perception of reality. Moreover, as time passes, and despite the state-of-the-art means to document the events resulting from the global pandemic, individual memories will start to fade or change due to external influence. The ICC report shows that human memory is not only subjective and prone to fading, but can also be easily distorted. This topic has not gathered a lot of interest among the arbitration community, until the creation of the Task Force on Maximising the Probative Value of Witness Evidence appointed by the ICC Commission on Arbitration.

The Report

The work undertaken by the Task Force resulted in the release of the Report on the Accuracy of Fact Witness Memory in International Arbitration on November 2020.

As the Report points out, there are several reasons explaining why the memory of fact witnesses is worth further consideration by the community. These reasons include in particular time and costs invested in witness preparation and the impact of witness evidence on the final decisions of arbitral tribunals.

Role of the witness

The Report identifies and describes several purposes of fact witness testimony, such as proving disputed facts, explaining documents, providing context and “telling the story”, as well as providing technical explanations. The report concludes that the accuracy of a witness’ memory does not have the same importance in all situations. For instance, when the witness provides technical background or information about a procedure then accuracy will have a lesser value as its testimony will be based on experience, education or knowledge. On the other hand, if fact witness testimony is about specific events the accuracy of memory is likely to be important.

Witness Memory in international arbitration

In any case, counsel and arbitrators should be aware that the memory of an honest witness can be easily distorted. The Report first identifies issues known in relation to the fallibility of human memory relevant to witnesses in other fields, such as criminal justice, where research has been more extensive. These issues include the impact which phrasing may have on responses to questions, how misleading information can interfere with the witness’ original recollection, or the fact that witnesses may be led to ‘remember’ entirely fabricated events (‘false memories’). These distortions may be caused simply by (well intentioned) interactions that commonly take place during the preparation and presentation of witness evidence. The Report analyses whether the same witness memory issues arise in international arbitration. A witness memory experiment was designed, with 316 participants working across a broad range of industries and roles. The results were similar to studies performed in the context of criminal law. This led to the conclusion that in some circumstances witness testimony may be less reliable than assumed by the parties involved in an arbitration, including the arbitrators themselves.


The Report provides a comprehensive set of recommendations for practitioners that would help to reduce distorting influences and their impact on witness evidence. The relevant means are addressed to in-house and external counsel as well as arbitrators. In particular, the recommendations for counsel focus on how to best conduct interviews (preferably witness by witness rather than as a group), assessing information relayed by witnesses, and the witness preparation ahead of a hearing. For arbitrators, the Report also pinpoints the ways to identify and asses the weight of distorting influences on witness memory.

The Report emphasizes the role of practitioners, including both counsel and arbitrators, in raising awareness on the functioning of human memory and on the strengths and weaknesses of the recounting process.


As a final remark, the Report recommends avoiding the conclusion that witness evidence is second best. It rather addresses existing concerns providing useful insight on how to minimise possible memory distortions or contamination. In this respect, the Report notes that a having a preestablished view of what of evidence is best does not seem to be justified or prudent.