Ciarb’s newly appointed Europe Branch chair, Dr Phillip Landolt
FCIArb, considers witness evidence in international arbitration.

Except for 'look sniff arbitrations' in the commodities sector, it is decidedly rare for an arbitration to be 'on the documents only'. This is true not just for common law inspired arbitrations, but also for civilian ones. It is rarely oral argument that keeps arbitration from being on the documents only. It is the virtually invariable resort to witnesses as a component of the evidentiary mix. Witness evidence is indeed a cardinal type of evidence in arbitration, perhaps vying only with documentary evidence in order of importance. 

But vis-à-vis documentary evidence, witness evidence presents remarkable complexities resulting in its probative value being starkly more variable. It varies extensively as a function of the characteristics of the person who is testifying, as a function of their perception of it, and as a function of what befalls that person after having apprehended it. This potential breadth of variation is moreover expanded further by factors external to the witness such as how counsel presents the witness evidence, how it is examined, and the arbitrator’s wherewithal to assess it. 

Since the assessment of witness evidence is so nettlesome, why does one even bother with it in adjudication and in arbitration in particular? The simple answer is that one of the most readily available means of finding out about virtually anything is to ask someone about it. To borrow from the classic wording of Rule 401 of the US Federal Rules of Evidence, witness evidence as a class is eminently capable of the 'tendency to make a fact more or less probable'. So since time immemorial witnesses have been resorted to to prove or disprove a case in dispute. 

Since witness evidence in international arbitration is a fact of life, the enquiry is when should, and how can, arbitration maximise its reliability, and the precision of assessing it. 

Reliability rules

In arbitration, there are a number of rules favouring the reliability of witness evidence. For example, witnesses will generally need to take an oath as to the truth, or include an affirmation of truthfulness at the end of a witness statement. Moreover, it is a generalised rule of practice that a witness may not attend another witness’ testimony until they themselves have been examined, and, if they are a party, they are examined before all other witnesses. Also, as reflected in Rule 4(7) of the IBA Rules on the Taking of Evidence in International Arbitration, in principle if a witness is requested to attend for examination and fails to do so their evidence is treated as inadmissible. Moreover, it is usual in arbitration to record the actual words of the witness’ oral testimony, which, in the form of a transcript, become an evidentiary record.

However many of the safeguards promoting the reliability of witness evidence which one finds in civil procedure before courts are in arbitration attenuated or missing altogether. The strictures on preparing witnesses are not applied. The reality in arbitration is that direct witness evidence is heavily coached often beyond the legitimate concerns of efficiency and ensuring that the witness is not overly affected by nervousness and unfamiliarity with the situation. The generalised use of witness statements in international arbitration favours a party’s control over its witnesses. At any rate, there is no strict prohibition on asking one’s own witnesses leading questions, and it happens routinely.

Despite the widespread use of oaths and statements of truth, penalties for untruthfulness are infrequent in arbitration, as contrasted with the serious consequences of perjury before most courts. Swiss arbitration is one of the rare instances where there can be criminal liability for wilful untruthfulness in arbitration testimony, but the enforcement of such liability is vanishingly rare. The absence of the formalism of court premises and attire also signals to witnesses a lack of solemnity requiring no unusual truthfulness. If a witness is aware of the confidentiality of arbitration this too dampens incentives to tell the truth.

Cross-examination

Perhaps the most significant reduction of incentives on witnesses to tell the truth is how opposing witnesses are examined. Arbitrators casually interrupt cross-examination. Breaks are ordered when examination starts to heat up. Increasingly decisions are made to conduct witness examination remotely, exclusively with an eye to costs and time savings, with diminishing offsetting concern for the ability to observe the witness and how they behave. Even where hearings are physical, witnesses are often placed so far away that it is difficult for arbitrators to assess their demeanour.

In arbitration it seems that there is a general acceptance of the legitimacy of business persons providing an overtly self-interested account of events, and they are not pressed to any extent where demeanour would register. 

This phenomenon may be attributed to two factors operating in international arbitration. For one, it is doubtless a by-product of the ethos of consent in arbitration. In a somewhat misplaced observance of this ethos, counsel and arbitrators are often reluctant to challenge witnesses and to expose untruthfulness. Secondly, it does seem that it is the product of the civilian law approach to witness examination.

Cross-examination is at the very heart of procedural rights in common law systems. The American Evidence Law professor John Henry Wigmore famously gushed that cross-examination is the "greatest legal engine ever invented for the discovery of truth." The cross-examination Wigmore was referring to is a highly developed and ramified system with rules of exclusion of evidence like the English and Commonwealth rule in Browne v. Dunn, and, crucially, the role of putting psychological pressure on witnesses to register their demeanour and gauge the firmness of their testimony. Witnesses are placed cheek by jowl with judges who scrutinise their every reaction. Judges do not interrupt cross-examination. Their judgments will generally address the quality of every witness’ evidence, and in particular their demeanour as a witness.

Questioning in Civilian Law

Wigmore’s statement tends to raise a chortle with civilian lawyers. They find it bombastic, and archly remark that Americans do fancy their engines. In civilian systems the judge leads the questioning. The judge will obligingly tender questions to the witness and patiently and passively listen to most any answer. Follow up questions to explore inconsistencies and a lack of clarity are a distinct rarity. Judges will admonish crass speculation, but are generally impervious to whether or not the witness had direct perception of what they are testifying to. 

Once the judge has finished asking all of his or her questions counsel will be invited to question the witness, first counsel who called the witness then opposing counsel.  The judge will be extremely begrudging in allowing follow up questions. The impression is generally that the witness has been given the opportunity to express themselves on a subject and whatever they say is self-delimiting. 

The judge will orally summarise the testimony for a reporter to enter into the minutes of the examination, which the witness will sign. Usually the questions asked, and, as a rule, the actual words the witness uses and their demeanour in using them, are entirely lost as elements of evidence. Judgments will almost never broach the quality of any witness’ evidence. 

The common law and its cross-examination distinctly favours the reliability of witness evidence and the judge’s ability to assess it. Civilian systems respond to the comparative unreliability of witness evidence by more frequent resort to other sources of evidence, in particular documentary evidence. But this leaves evidentiary gaps, vis-à-vis what prevails in common law systems. In civilian judgments these tend to be filled with more prominent use of evidential inferences, and reliance on the burden of proof to settle evidentiary questions.

Arbitration approaches

In arbitration the relaxation of common law rigour in cross-examination generally entails a depreciation in the reliability of witness evidence.  Perhaps equally worrisome, a party’s reliance on the reduced evidentiary standards in arbitration can be severely punished where, contrary to this practice, the arbitrators unexpectedly apply a rigorous common law standard. The award of the highly distinguished arbitrators that recently Knowles J. in Nigeria v. PID refused to enforce was powerfully, perhaps conclusively, influenced by the failure of the respondent’s counsel in cross-examination to challenge factual assertions in the claimant’s quantum claim.  It may well have been that, in the informal context of arbitration, the respondent expected a less unitary and purist approach.

It seems by consequence highly advisable for arbitrators at the outset to make clear what their approach will be to witness evidence in view notably of the degree to which it is likely to play a prominent role in the arbitration. In doing so they should account for the parties’ expectations as to maximising its reliability, and the precision of assessing it. On the whole, there should be heightened concern to ensure the reliability of witness evidence by incorporating appropriate procedural mechanisms, in particular closer adherence to the model of common law cross-examination. There should especially be a sensitivity on all sides to the evidentiary impact of witness demeanour, and awards should address this and the general quality of each witness’ evidence.