The right to an independent and impartial judge in the determination of one’s civil rights or a criminal charge against them is recognized in all advanced legal systems. It is also often said that justice must be done, and must be seen to be done, because the rule of law depends on public confidence in the administration of justice.

English law seeks to uphold these principles primarily through the law of apprehended bias. Judges cannot sit where a fair minded observer [FMO] would conclude that there was a real possibility that the judge was biased. In an article published last year in the Civil Justice Quarterly, we argue that test for apprehended bias fails to effectively promote either principle.

It is widely accepted that the test for the apprehended bias is itself susceptible to cognitive biases, and also risks undermining public confidence, because it involves judges judging their own capacity to decide cases impartially, albeit through the eyes of a fictional observer who is also created by the judge. In the paper we use the deliberately provocative, but in our view justified, metaphor of the puppet master asking the puppet what they think of their master.

However, we argue that there is a more fundamental flaw in the law of apprehended bias, which is a lack of institutional competence on the part of those applying it. Even if judges were capable of avoiding their own cognitive biases when reviewing their own position, they are ill-equipped to decide whether they should be disqualified by reference to the FMO, or by reference to objective risks of partial decision making. The reason is that these metrics are not legal in nature. Neither involves any technical legal analysis in assessing whether the judges should not sit, and yet the task of deciding this question is made solely by lawyers.

By contrast, the meaning of legal bias is axiomatically a legal question. Defining legal bias raises fascinating questions about the role of the judge. For example, should a judge ever draw on their own life experiences when deciding a case? This might lead to better judging in some cases, especially where the judge has some insight into the parties’ experiences, but because any person’s experience is necessarily limited, it might also lead to error, or even bias.

The traditional view, at least in English law, has been that justice is blind in the sense that judges should try to exclude, from their consideration, everything other than the evidence before them, and apply the law as they understand it to the facts as they find them: Locabail (UK) Ltd v Bayfield Properties Ltd [2000] QB 451. In short, the law presupposes judges can avoid the influence of anything other than the evidence when deciding cases before them. From the perspective of behavioural psychology this is an impossible objective, and even the courts have readily acknowledged that no judge is a tabula rasa and cannot eliminate entirely the influence of their own life experiences from their decision making. As a result, the law of apprehended bias has focused on ensuring judges can keep an open mind, so as to prevent the influence of extraneous factors, and the cognitive biases they lead to, from becoming a real danger of bias. Yet judges have no expertise in relation to the risks of cognitive biases influencing decision making, nor any special knowledge of public attitudes as to when they think this risk of bias is unacceptably high. There are a number of psychological studies that call into question key assumptions on which the law of apprehended bias is based, and whether the court’s application of the FMO test reflects actual public opinion remains unknown. Commentators and even some judges have acknowledged that the courts have created an idealized observer endowed with knowledge and attributes that many would struggle to attain. This is partly because of a judicial desire to ensure that policy considerations related to the functioning of the justice system are also taken into account in disqualification decisions.

We have begun conducting representative surveys in both the UK and Australia to assess when the public believes judges should be disqualified. That research, which we aim to publish later this year, suggests that the public thinks judges should be disqualified in a far broader range of cases than is required under the current law.

For all of these reasons we propose a new framework for the law of apprehended bias, based on the principle that the law should not be determined exclusively by lawyers, but instead be developed through a policy-making process that expressly balances reliably measured public opinion, legal policy considerations and the science on human decision making. This policy process will also allow for a dialogue between the public, the legal community and psychologists as to what is possible, and desirable, in framing the law of apprehended bias, and ultimately the creation of a code regarding judicial disqualification. We also outline novel tests for dealing with borderline cases which would end the need for judges to decide their, or their colleagues’ ability, to decide a case impartially.

Andrew Higgins is Associate Professor of Civil Procedure at the Faculty of Law and Mansfield College, University of Oxford.

Inbar Levy is a Lecturer in Law at Melbourne Law School, University of Melbourne.