Assessing credibility and the limits of demeanour – Khan v GMC  EWHC 374 (Admin)
Last year in Dutta v GMC  EWHC 1974 (Admin), Warby J found the Tribunal’s reasoning contained at least three fundamental errors. First, the Tribunal had approached the resolution of the central factual dispute by starting with an assessment of the credibility of a witness’s uncorroborated evidence about events ten years earlier, and only then went on to consider the significance of unchallenged contemporaneous documentation. Secondly, the Tribunal’s assessment of the witness’s credibility had been based largely if not exclusively on her demeanour when giving evidence. Thirdly, the way the Tribunal had tested the witness evidence against the documents involved a mistaken approach to the burden and standard of proof. In a case that involved a conflict between a witness’s recollection and unchallenged documentation, Warby J held that it was an error of principle to ask “do we believe [the witness]?” before considering the documents, and stressed that reliance on a witness’s confident demeanour was a discredited method of judicial decision-making.
In Khan the Registrant faced allegations of sexual misconduct made by three complainants A, C and D. His defence was that the allegations had been fabricated or exaggerated.
Notwithstanding Warby J’s trenchant comments and ruling in Dutta, in Khan the Tribunal declared witness D, who had been the first witness to complain, to be “credible and consistent” before it considered any of the evidence she had given. Unsurprisingly, given this finding, the Tribunal preferred her account to that of the Registrant. However, the Tribunal had heard evidence from two other witnesses who had flatly contradicted witness D’s evidence that one or other of them had said to witness D that the Registrant had made inappropriate approaches to them as well. The Tribunal failed to deal adequately or at all with this conflict and did not appear to consider how it might have affected witness D’s credibility.
Similarly, the Tribunal stated that it found witness A to be a “confident, credible witness” before considering any of the allegations in detail. It did not deal, inter alia, with a direct conflict of evidence between her and a police officer to whom she had made a complaint.
As to witness C who had accepted lying on oath in related proceedings before the Employment Tribunal and fabricating the contents of an anonymous letter that had implicated the Registrant, the Tribunal stated that it had first considered her credibility and, specifically, the extent to which it was undermined by her admissions. It had assessed her “demeanour” in making its credibility assessment and found she had been “adamant” that she had been truthful in her evidence.
Knowles J held that, on examination, the Tribunal’s reasons for concluding that C was “credible” showed that it had fallen into the precise trap which Warby J in Dutta had warned against. Effectively, it had begun its analysis of her evidence by asking “Do we believe her?”
Further, the Tribunal’s language showed that its reasons were based in significant part on the twin fallacies that “the more confident another person is in their recollection, the more likely it is to be accurate” and “because a witness has confidence in his or her recollection and is honest, evidence based on that recollection provides any reliable guide to the truth”. Knowles J reiterated that “Reliance on a witness’s confident demeanour is a discredited method of judicial decision making”, all the more so in the case of a witness who had admitted lying on oath on a previous occasion.
It is clear that in both Dutta and Khan there was significant evidence that undermined the witnesses, which the Tribunals simply failed to deal with appropriately. There is now a wealth of persuasive first instance authority from various divisions of the High Court on the inappropriateness of demeanour as a tool with which to assess evidential cogency.
However, what of those cases, a very significant number of which will involve allegations of sexual assault, where there is no evidence other than that of the complainant and the Registrant and no extraneous evidence casting doubt on credibility of either? In these cases, Tribunals are now faced with authoritative ‘myth-busting’ guidance not to be influenced by preconceptions as to how victims of a sexual assault might behave either in the aftermath of an incident or when they come to give evidence about it. At the same time following Dutta and Khan, Tribunals should now understand that demeanour, and in particular the strength or vehemence of the way in which a witness gives their account, is not a reliable pointer to credibility. In those circumstances, a Tribunal may be forgiven for asking how is it to determine whether or not a complainant or a Registrant is credible?