Pump Court Chambers

Remote hearings and witness evidence

Blog 29th April 2020
Ezra MacDonald

Remote hearings come with particular difficulties. One obvious difficulty is that the parties cannot see each other as well as they might in the courtroom. But how much of a problem is this?

If the demeanour of witnesses (or indeed parties) is relevant to an issue to be decided by the Court, then restrictions on the Court’s ability to evaluate witness demeanour are significant – and might prevent a fair trial.

If the demeanour of witnesses is irrelevant (or, worse, outright misleading), then restrictions on the Court’s ability to evaluate witness demeanour simply don’t matter. They should not be an obstacle to proceeding with a remote hearing (although there may be other reasons for an in-person hearing).

Rather unhappily, the case law on this issue is a) inconsistent between jurisdictions – specifically as between criminal and civil cases, and – even worse – b) inconsistent within jurisdictions, with conflicting cases in both civil and family jurisdictions.

This is a real problem, because a live issue in a great number of cases is whether they can fairly be dealt with remotely. It needs to be resolved. My view is that the correct approach is the one taken in the civil courts (following R (SS) cited below). It must be the correct approach because it is the only approach informed by science. I discuss some of the relevant research at the end of this post.


Civil and family cases

In R (SS) v Secretary of State for the Home Department [2018] EWCA Civ 1931 Leggatt LJ had some fairly trenchant observations to make about witness demeanour and its importance (or lack of importance) when it comes to assessing credibility.

[46] No doubt it is impossible, and perhaps undesirable, to ignore altogether the impression created by the demeanour of a witness giving evidence. But to attach any significant weight to such impressions in assessing credibility risks making judgments which at best have no rational basis and at worst reflect conscious or unconscious biases and prejudices. One of the most important qualities expected of a judge is that they will strive to avoid being influenced by personal biases and prejudices in their decision-making. That requires eschewing judgments based on the appearance of a witness or on their tone, manner or other aspects of their behaviour in answering questions. Rather than attempting to assess whether testimony is truthful from the manner in which it is given, the only objective and reliable approach is to focus on the content of the testimony and to consider whether it is consistent with other evidence (including evidence of what the witness has said on other occasions) and with known or probable facts. (emphasis added)

In the Family Division of the High Court, the point was applied by Macdonald J in Cumbria County Council v S, E, R[2019] EWHC 2782 (Fam):

  1. Within the context of the foregoing legal principles, this court must bear in mind that the assessment of the credibility and reliability of the parents should coalesce around matters including the internal consistency of their evidence, its logicality and plausibility, details given or not given and the consistency of their evidence when measured against other sources of evidence (including evidence of what the witness has said on other occasions) and other known or probable facts. The credibility and reliability of that parent should not be assessed simply by reference to their demeanour, degree of emotion or other aspects of their presentation. (emphasis added)

Back in the Civil Division of the Court of Appeal, in Staechelin, Paisner, McCaffrey v ACLBDD Holdings Limited & ors [2019] EWCA Civ 817, Lewison LJ considered Yaqoob v Royal Insurance (UK) Ltd [2006] EWCA Civ 885 and remarked:

  1. What seems to have been the deciding factor for the trial judge was Mr Yaqoob’s demeanour in the witness box. At [36] Chadwick LJ held that the finding was flawed because it failed to deal with the real issues of credibility, namely (a) the conflict of evidence (b) the agreed profile of the arsonist and (c) the low level of stock on the premises. In short, the judge had not taken advantage of the benefits of seeing and hearing the witnesses. Chadwick LJ concluded at [38]:

“If, as I have sought to explain in the present case, the judge has not taken proper advantage of that opportunity — by failing to make findings of fact which were essential, by failing to address the question of credibility and by failing to analyse and give proper weight to the necessary conclusions to be drawn from the forensic evidence as to the profile of the perpetrator — then it cannot be enough for this court simply to say, “Oh well, the judge believed the witness and so must we”.”

  1. It is, to my mind, of critical importance that the trial judge based his evaluation on the demeanour of Mr Yaqoob in the witness box. For the reasons that Leggatt LJ explained in R (SS) (Sri Lanka) v Secretary of State for the Home Department [2018] EWCA Civ 1391a conclusion simply based on the demeanour of a witness is not built on a solid foundation. (emphasis added)

It is perhaps significant that in S, E, R as in Staechelin the Court has qualified what was said in R (SS) by the addition of the word “simply”. With respect, I think this is an error. The point in R (SS) is that “to attach any significant weight to [impression created by demeanour] risks making judgments which at best have no rational basis . . .”

In Liverpool City Council v M, F, C [2018] WL 06249888 (another first-instance decision), the Court had been invited to make findings about allegations of serious sexual abuse of a physical nature. There was disagreement as to whether the (by that time adult) complainant should give evidence by way of screens or video link. The subject of the allegations said that the Court’s ability to assess the credibility of the witness would be compromised if the Court were unable to assess the witnesses’ demeanour when giving evidence. HHJ Greensmith noted that “[t]he relevance of demeanour as an indicator of credibility is questionable”, and that the Judicial Colleague’s teaching was that judges should be “very circumspect” about the value of demeanour. The Court also referred to R (SS). Directions for evidence by way of video link were given.

In Auliffe and ors v Ellis [2019] EWHC 1427 (QB) (an appellate decision), Baker J remarked (perhaps slightly opaquely):

  1. Demeanour in the witness box is indeed understood to be no general indicium of honesty. But that is not to say that judging issues of primary fact involves no element of reading witnesses as individuals, and one aspect of that is, or at least can be, evaluating the meaning or significance, if any, of variation in a witness’s demeanour as between different topics or particular questions. Nor does treating demeanour as, in general, an unreliable guide to reliability mean that reading a written transcript is always, or even usually, as good as being at the trial.

Again, that is to dilute the observations of the Court of Appeal in R (SS). What could be involved in “reading witnesses as individuals” other than assessing demeanour? The observation in R (SS) is that to attach any significant weight to demeanour risks making judgments which “at best have no rational basis”.

Conversely, in Re P (A Child: Remote Hearing) [2020] EWFC 32, the Court was asked to consider whether a case involving “a particular form of child abuse which requires exquisite sensitivity and skill on the part of the Court”. The President of the Family Division commented that

“. . . it is a crucial element in the judge’s analysis for the judge to be able to experience the behaviour of the parent who is the focus of the allegations throughout the oral court process; not only when they are in the witness box being examined in chief and cross-examined, but equally when they are sitting in the well of the court and reacting, as they may or may not do, to the factual and expert evidence as it unfolds during the course of the hearing.”


Criminal cases

In R. v. Popescu [2011] Crim L R 227, CA, the Court considered whether the jury should have been given the transcript of the complainant’s evidence and said this:

  1. Secondly, if the transcripts are given to the jury, we suggest, first, that the judge must warn the jury then and there to take care to examine the video as it is shown, not least because of the importance of the demeanour of the witness in giving evidence. (emphasis added)

More generally, the Crown Court compendium refers at various points to the importance of demeanour – e.g. where discussing hearsay evidence, noting that the jury should be warned about the limitations of that evidence, including “the inability of the jury to assess the demeanour of the witness . . .”

Moreover, the judge has the option of suggesting that although the jury may take notes (during evidence), “it would be better not to take so many notes that they are unable to observe the manner / demeanour of the witnesses as they give their evidence.”

The one point in the Compendium at which demeanour is suggested to be misleading is when it comes to sexual offences:

“The experience of judges who try sexual offences is that an image of stereotypical behaviour and demeanour by a victim or the perpetrator of a non-consensual offence such as rape held by some members of the public can be misleading and capable of leading to injustice . . .”

Lastly, in R v D [2014] 1 LRC 629, a first instance decision, the judge considered whether the defendant should be permitted to testify in her own defence without removing her niqab. HHJ Murphy stated that “it is essential to the proper working of an adversarial trial that all involved with the trial – judge, jury, witnesses, and defendant – be able to see and identify each other at all times during the proceedings.” He decided that the defendant would need to remove her niqab in order to give evidence. At paragraph 59 he said this:

“It is unfair to ask a juror to pass judgment on a person he cannot see. It is unfair to expect that juror to try and evaluate the evidence given by a person whom she cannot see, deprived of an essential tool for doing so: namely being able to see the demeanour of the witness; her reaction to being questioned; her reaction to other evidence as it is given.” (emphasis added)



It has been known for some time that people are not good lie detectors (DePaulo, B. M. 1994). Their ability does not necessarily improve with practice: for instance, one study found that federal law enforcement offers working in jobs which require attempts to detect deceit were no better at detecting deceit than untrained, undergraduate students – they only thought they were (DePaulo, op. cit.)

The point is plainly made in Jeremy Blumenthal’s 1993 discussion on the topic (available here), which concludes – with some cogency – that “[i]t is unforgiveable that the legal system deliberately ignores demonstrated, relevant findings about demeanor evidence and wilfully adheres to an ineffectual traditional approach.”

A comprehensive meta-analysis by Bond & DePaulo (Bond & Depaulo ) synthesised results from 206 documents and 24,483 “judges” (i.e. test subjects) and found that people achieve an average of 54% correct lie-truth discrimination – which only increased marginally for professional lie-catchers. In other words, just a little better than a coin toss. Worse still, the ability deteriorates with age (Curci et al., infra)

The issue is discussed in wide-ranging detail by Curci et al. (Curci et al., 2009), available here. Rather problematically, the authors comment that “judges and jurors evaluate witness evidence based upon categories which correspond to what laypeople usually consider as indicators of turhtful / deceptive criteria.” The authors comment that “[p]eople rate themselves as sufficiently expert at identifying lies from the interlocutor’s physiological pattern and expressive behaviour, but the laypeople’s ability at lie/truth discrimination based upon non-verbal signals has been demonstrated as being only slightly above chance . . . the basis for the legal evaluation of witness evidence across jurisdictions is experiential and, as such, mainly unwarranted.” (emphasis added).

Arguably, jurors should be given a direction that this is the case, and judges should direct themselves of the fact (much in the manner of a Lucas direction, i.e. that a witness may lie for many reasons . . .)



The judicial guidance is not merely inconsistent, it is starkly inconsistent. This observation has already been made (in slightly less strident terms) by Lady Hale in her 2019 address on religious dress, given to the Woolf Institute (available here): the Judicial College guidance cautions against evaluating credibility from demeanour in civil cases, but advises judges to be “particularly careful to point out that [wearing the veil] might impair the court’s ability to evaluate the reliability and credibility of the wearer’s evidence . . .” in criminal cases.

For the reasons I have set out above, I take the view that the approach should be harmonised across jurisdictions and should follow the very sensible guidance given in R (SS). It is the only guidance which is informed by, and consistent with, the scientific research. It is the most likely to be correct.

Whatever the obstacles to proceeding with a remote hearing are, the hypothetical difficulty of assessing a witness’s credibility from her demeanour should not be one of them.


This article on ‘Remote hearings and witness evidence’ was written by Ezra Macdonald and Simon Purkis if you would like any further information on instructing Ezra or Simon  have any other queries please contact our clerking team through our switchboard 020 7353 0711 or via email.


Blumenthal, J. A. 1993 “A Wipe of the Hands, A Lick of the Lips: the Validity of Demeanor Evidence in Assessing Witness Credibility”, 72 Neb. L. Rev.

Bond CF Jr, DePaulo BM. Accuracy of deception judgments. Person Soc Psychol Rev. (2006) 10:214–34. doi: 10.1207/s15327957pspr1003_2

Curci, A. et al. 2019 “Accuracy, Confidence, and Experiential Criteria for Lie Detection Through a Videotaped Interview” Front. Psychiatry 22 January 2019, doi: 10.3389/fpsyt.2018.00748

DePaulo, B. M. 1994 “Spotting Lies: Can Humans Learn to Do Better?”, Current Directions in Psychological Science3(3): 83 – 85.

Shortlist close
Title Type CV Email

Remove All


Click here to email this list.