Ways to curb expert bias

September 1st, 2017

By Dr Izette Knoetze-le Roux

The expert’s function, is an exceptional one: To provide assistance to courts in cases where the court is unable to, because of lack of specialised knowledge.

Expert evidence refers to evidence, which is admissible because it is relevant and of probative value in relation to matters in issue in the case and as such is a particular category of admissible evidence (George Barrie and Bertus de Villiers ‘Revisiting the adversarial approach of dealing with expert evidence: The treatment of expert witnesses by the state administrative tribunal of Western Australia’ (2017) 1 Tydskrif vir Suid-Afrikaanse Reg 59). This type of evidence is permissible if it is relevant and of probative value. Where expert evidence is admitted it is admissible because it has probative value.

It will have probative value because there are issues before the court – which requires specialist knowledge or experience for their resolution – and evidence is being adduced from a person who has specialist knowledge or experience appropriate to those issues.

The dangers associated with the use of expert evidence is often limited to bias, namely, the risk that the expert, hired by the prosecution or the defendant, will report results that are favourable to the party who instructed him or her and paid his or her fee (Joëlle Vuille ‘Admissibility and appraisal of scientific evidence in continental European criminal justice system: Past, present and future’ (2013) 45 Australian Journal of Forensic Sciences 389).

David Bernstein (‘Expert Witnesses, Adversarial Bias, and the (Partial) Failure of the Daubert Revolution’ (2008) 93 Iowa Law Review 451) notes that expert testimony is vulnerable to ‘adversarial bias’, which refers to witness bias that arises because a party to an adversarial proceeding retains experts to advance its cause.

Three sources of adversarial bias

Adversarial bias has three sources, namely –

  • conscious bias;
  • unconscious bias; and
  • selection bias (Bernstein (op cit) at 452).

Conscious bias, is evident when ‘hired guns’ adapt their opinions to the needs of the client and legal practitioner who hires them. Unconscious bias exists across various categories of expertise, but it is an especially persistent problem with regard to testimony by forensic scientists (Bernstein (op cit) at 455 – 456). As a result, the forensic expert’s unconscious bias can easily have an impact on the conclusions arrived at. Selection bias, refers to the fact that the experts retained by a party will not represent a random sampling of expert opinions. Rather, they will represent the perspective the legal practitioners want to present at trial.

In order to address the difficulties of adversarial bias, a number of proposals are discussed below.

Will tighter admissibility requirements of expert witnesses reduce bias?

  • Position in England

In the matter of Helical Bar PLC and Another v Armchair Passenger Transport [2003] EWHC 367 (QB) the High Court in England compiled a list of principles in relation to the independence and impartiality of expert witnesses. The court held that where an expert has an interest in the outcome of the case, the judge must decide ‘whether [the] expert should be permitted to give evidence’. The decision on admissibility is a question of ‘fact and degree’ and depends, in part, on whether the proposed expert ‘is aware of their primary duty to the Court … and willing and able, despite the interest or connection with the litigation or a party thereto, to carry out that duty’. When the expert’s evidence is admitted, the court will consider how ‘an interest which is not sufficient to preclude him from giving evidence’ will affect the weight to be accorded to the expert’s opinion.

  • Position in Canada

The Canadian Supreme Court’s decision of White Burgess Langille Inman v Abbott and Haliburton Co. 2015 SCC 23 tightened the admissibility requirements of expert witnesses. The Supreme Court confirmed that expert witnesses owe a special duty of impartiality at common law to provide ‘fair, objective and non-partisan assistance’ to the trier of fact.

Hunt notes that this duty is underpinned by the three concepts of impartiality, independence and absence of bias (CDL Hunt ‘The expert witness’s duty of impartiality in Canada: A comment on White Burgess Langille Inman v Abbott and Haliburton Co.’ (2016) 20 The International Journal of Evidence and Proof 73). First the opinion must be ‘impartial’ in the sense that it reflects an objective assessment of the questions, secondly the opinion must be the ‘product of the expert’s independent judgment, uninfluenced by who has retained him or the outcome of the litigation’ and finally the opinion must be unbiased in the sense that it does not unfairly favour one party over another. On the question of admissibility, a threshold eligibility requirement for the admission of expert evidence was imposed (Hunt (op cit) at 72).

It further held that courts should exclude such testimony if the expert lacks sufficient independence and impartiality. An expert will be qualified to testify only if he is aware of and willing to carry out their duty of impartiality to court. In addition, the trier of fact continues to act as gatekeeper as it will assess the probative value of the proposed expert’s evidence and weigh it against the potential for prejudice. It will only be admitted where the probative value exceeds the potential for prejudice.

Expert opinion evidence must pass the two-stage Mohan framework (R v Mohan [1994] 2 SCR 9). The first stage relates to the expert’s proper qualifications. This entails that a properly qualified expert must be both aware of and willing to carry out his duty to court to be impartial and independent. The expert’s attestation ‘recognising and accepting the duty’ will generally be sufficient. The opposition might challenge this by showing a ‘realistic concern’ that the expert is unable or unwilling to comply with this duty. Should this occur, the proponent of the evidence must prove on the balance of probabilities that the expert understands and intends to carry out his duty to the court; failure to do so renders the evidence inadmissible. On assessment, the courts will consider the nature and extent of a party’s interest in the litigation. Disqualification should only occur in ‘very clear cases in which the proposed expert is unable or unwilling to provide the court with fair, objective and non-partisan evidence.’

If this stage of Mohan is passed, the proponent of the evidence must satisfy the court that the probative value of the proposed expert’s evidence outweighs its potential for prejudice. Here the judge acts as a gatekeeper, holding a ‘residual discretion to exclude evidence based on a cost-benefit analysis’.

  • Position in Australia

In FGT Custodians (Pty) Ltd v Fagenblat [2003] VSCA 33 at para 26 the Supreme Court of Victoria, Court of Appeal rejected the argument that an interest in the case could implicate the competency of an expert witness to testify. The court noted that for ‘at least 100 years if not far longer’, the remedy for the criticism of an expert witness ‘has not been seen in denying the right of such witnesses to give evidence; rather it has been seen in devising court rules and protocols which will ensure that experts will try to be independent and that courts will not unnecessarily suffer the opinions of experts who may be thought to be in one camp or the other.’ The court expressly declined to lay down a new common law rule that would exclude expert witness evidence unless it passed a threshold assessment of its impartiality.

  • Position in South Africa

An expert witness should remain objective despite the fact that he or she is called by a party to testify in support of the latter’s case. This principle was adopted by South African courts in the matter of Stock v Stock 1981 (3) SA 1280 (A) and recently reaffirmed by the Supreme Court of Appeal (SCA) in the matter of Jacobs and Another v Transnet Ltd t/a Metrorail and Another 2015 (1) SA 139 (SCA). In this matter the SCA had to consider how to approach conflicting expert opinions. Majiedt JA (with Navsa ADP, Saldulker JA, Swain JA and Zondi JA concurring) noted at para 15 as follows:

‘It is well established that an expert is required to assist the court, not the party for whom he or she testifies. Objectivity is the central prerequisite for his or her opinions. In assessing an expert’s credibility an appellate court can test his or her underlying reasoning and is in no worse a position than a trial court in that respect. Diemont JA put it thus in Stock v Stock: “An expert . . . must be made to understand that he is there to assist the court. If he is to be helpful he must be neutral. The evidence of such a witness is of little value where he, or she, is partisan and consistently asserts the cause of the party who calls him. I may add that when it comes to assessing the credibility of such a witness, this court can test his reasoning and is accordingly to that extent in as good a position as the trial court was.”’

Is conferral of experts prior to hearing and concurrent evidence by experts during hearing the viable solution to expert bias?

Australian Courts have been experimenting with two principal techniques to better assess expert evidence and to simplify the process of dealing with expert evidence. The techniques are conferral of experts prior to hearings and concurrent evidence by experts during hearings.

The aim is to identify the areas of agreement, reduce the scope of disagreement, identify areas of disagreement and clarify the reasons for disagreement. These processes are viewed as reducing the adversarial nature of expert evidence, reducing time required for expert evidence, creating an immediacy of peer review and ensuring a better informed court or tribunal (Bertus de Villiers ‘Accessibility to law: Adjusting court proceedings to the modern era – novel practices and procedures from down under’ (2016) 14 New Zealand Journal of Public and International Law 229).

Conferral of experts

Expert conferral provides a potential mechanism to better identify issues in dispute, save time and clarify complex issues for clients, lawyers and the courts. Although expert conferral does not necessarily reduce the number of issues in dispute, it can provide a basis for the reasons for disagreement to be clearly articulated: Thereby assisting the judiciary with examination and weighing of expert opinion.

Concurrent expert evidence

According to De Villiers (op cit) at 242, concurrent expert evidence is a process during a hearing whereby experts, from the same or related disciplines, present evidence to a court or tribunal during a joint session of those experts. In general, the experts are called together, sworn or affirmed and given the opportunity to answer the same questions, to comment on each other’s replies, to enter into a dialogue with each other and to put questions to each other.

The process of concurrent evidence is often preceded by conferral of experts where the joint report produced during a conferral sets the agenda for the concurrent evidence and the examination of experts.


Courts should reiterate that an expert witness’s professional duty is not to the party that pays him or her, but to the court.

The Australian process of expert conferral entails that experts meet without their clients or legal representatives in attendance and without the experts taking instructions or discussing the content of the joint report with clients or legal representatives prior to it being filed in the court or tribunal. This ensures that experts are treated as witnesses of the court and not advocates of a client, clients can easily feel alienated since they may lose control of the process. The joint report produced by the experts is usually handed to the court and parties.

The process of concurrent evidence, as noted by De Villiers (op cit) at 243 serves to encourage a less adversarial atmosphere during the trial and emphasises actual disagreements and reasons for those disagreements. It further serves to contribute to more collegial interactions between experts and assist the court and parties to focus on the real issues in dispute.

The concurrent giving of evidence by experts tends to highlight that experts have an obligation to assist the court rather than be an advocate for the client. The two processes adopted by the Australian courts provide courts with an opportunity to assess and weigh the opinions more effectively.

These processes have contributed to time being saved, changing the setting of hearing to less adversarial and more collegial and to litigants in person being able to better conduct and manage their cases since the issues are clearly identified and experts give evidence concurrently.

It follows that the abovementioned Australian approach offers a viable mechanism for diminishing expert bias.

Dr Izette Knoetze-le Roux LLB LLM (cum laude) LLD (UFS) is a Legal Researcher at the National Office of Legal Aid South Africa in Johannesburg.

This article was first published in De Rebus in 2017 (Sep) DR 37.

De Rebus