This week I’m looking at a case from the Chancery Division of all places, Instrument Product Development Ltd v WD Engineering Solutions Ltd.
Why, I hear you ask? Well, it caught my eye because the judgment starts by talking about a noted 1981 study, Role of schemeta in memory for places (was it just me that had to Google the meaning of “schemata”?) by psychologists William Brewer and James Treyens, and went on:
“The study demonstrates an aspect of the fallibility of memory. We do not store memories as images, like a photo album, to be revisited in detail at a later date. We recreate the image every time we recall it, combining the details of what we do recall with our expectations of what we should recall. The process is automatic, and done without conscious realisation that it is taking place.
That issue is at the heart of this case. In the Brewer and Treyens study, different witnesses had different recollections of the same room that they had seen only seconds before. In this case, two witnesses have critically different recollections of the same telephone conversation held in March 2017 to which they were the only parties.”
It’s certainly not the first time that the courts have considered the reliability (or otherwise) of the memories of witnesses, and we’ve also blogged on the topic before. However, the case provides a useful summary of the relevant principles, and the opportunity to consider those in the context of construction adjudication.
Instrument Product Development Ltd v WD Engineering Solutions Ltd
It all started some five years before the case came to court.
The parties’ dispute concerns the design and manufacture of certain prototype props for Nespresso. Between December 2016 and February 2017, IPD designed and made prototype props, with WDES making some elements of them. At the end of February and into early March 2017, discussions took place regarding a joint venture arrangement for the work for Nespresso, which was focused on its store in Cannes. This culminated in a telephone conversation on 6 March 2017 between Mr Paget of IPD and Mr Beale of WDES. The following day, on 7 March 2017, Mr Paget emailed Mr Beale setting out the terms of what had been agreed.
Subsequently, the parties discussed a further pilot at a store in America. This was followed by other requests, including for Nespresso’s headquarters in Lausanne, and pilots in Mexico, Portugal and the UK. Another key event appears to have been a meeting with Nespresso in Lausanne in May 2017, which ultimately led to a much bigger order from it. By December 2017 it was the start of the end of the parties’ relationship.
The main issue for the court was the scope of the original agreement, reached in early March (the 6 March telephone call and 7 March email). Did it cover all future supplies of the prototype props that IPD was designing or was it focused on work for the store in Cannes?
The judge, Mr Richard Farnhill sitting as a Deputy High Court judge, started his judgment by commenting on the witnesses. (I find these comments on witnesses normally provide a good idea of the way the case is likely to go.) Mr Paget was described as a “helpful witness” and Mr Beale as claiming to have almost no recollection of areas that “might reflect poorly on his conduct in this matter or WDES’s case”.
The judge then reviewed the relevant authorities on the interpretation of oral contracts, and distilled five principles, which are helpful to set out:
“(i) In interpreting an oral contract, the parties’ subjective understanding about what they were agreeing is relevant and admissible evidence.
(ii) What is critical is their understanding at or immediately after the point at which contract is entered into. That is the only sensible reading of Lord Neuberger’s reference to their ‘contemporary understanding’.
(iii) Later statements and actions are much less reliable indicators of what the parties understood to have been agreed. Just as memory is affected by the process of preparing for trial, it is affected by seeing how a transaction works out in practice, the ‘interference from new experiences’ referred to by Leggatt J.
(iv) Subsequent conduct and statements may be relevant to questions of variation and estoppel, just as they may be in the case of written agreements. In those cases, the analysis is principally, but not exclusively, an objective one. The focus is on what the parties said and did, more than on what the parties thought.
(v) To the extent that it exists, documentary evidence of what was said in meetings and conversations will almost inevitably be a more reliable guide than the witnesses’ unaided recollections.”
The judge then applied these principles to the facts, concluding that the 7 March email was intended to record the parties’ agreement and it was the “best evidence of its terms”. Those terms only applied to Cannes, not to any future rollout. However, when prototype props for further pilots were requested, the parties handled them in a way that showed they intended them to be covered by their agreement, such that the original agreement was varied. This happened either expressly or by conduct.
The judge found that the email sent the day after the meeting in March 2017 was far more reliable evidence of what the parties had agreed than the recollections of the witnesses, which is consistent with his fifth principle set out above. That same principle equally applies in adjudication and other forums, whether interpreting oral contracts or other oral agreements parties may make, and most tribunals will employ the maxim “contemporaneous evidence is king”.
The judge also referred to something called “the valley of the norm”, which is derived from Noise by Kahneman, Sibony and Sunstein. This is where a witnesses’ recollection of an event is influenced by subsequent events. Applied in this case, the parties attached more weight to the meeting in Lausanne after it had happened than they had in the run-up to it taking place (Nespresso placed a large order after the meeting). This meant the judge placed only limited reliance on their recollections of the meeting.
However, what about where there is little or no contemporaneous evidence and a tribunal needs to decide between the evidence of two or more witnesses?
In my experience, this is relatively uncommon in adjudication, particularly in a digital age where much is recorded in writing, whether that be in emails, WhatsApp messages, etc. However, it is not unheard of and adjudicators have to decide which of the witnesses’ versions of events they prefer, possibly by testing it against inherent probabilities, or by asking the witnesses questions in writing or at a meeting. Deciding between the witnesses on the basis of which party bears the burden of proof should be a last resort.
Where an adjudicator prefers the evidence of Witness A, Witness B shouldn’t assume that the adjudicator has necessarily formed the view they were lying, or even telling some little lies. It may well simply be the case that the adjudicator considers that Witness A’s recollection of events might be more reliable then Witness B’s.
I will leave you with the words of Leggatt J (as he was then) in Gestmin SGPS SA v Credit Suisse (UK) Ltd, which I think sums up very well why the memory of witnesses in adversarial proceedings can so often be polarised:
“Memory is especially unreliable when it comes to recalling past beliefs. Our memories of past beliefs are revised to make them more consistent with our present beliefs. Studies have also shown that memory is particularly vulnerable to interference and alteration when a person is presented with new information or suggestions about an event in circumstances where his or her memory of it is already weak due to the passage of time.”