In search of an evidence-based test for judicial bias

The Canadian principle of judicial bias has remained static for the four decades since the Supreme Court of Canada decision in Committee for Justice v. The National Energy Board.  The Supreme Court’s description of bias is rooted in the 20th-century jurist’s “reasonable man” mythos and emergent mid-century theories about the mind:

… the apprehension of bias must be a reasonable one, held by reasonable and right minded persons, applying themselves to the question and obtaining thereon the required information. … Would he think that it is more likely than not the Decision maker, whether consciously or unconsciously, would not decide fairly.[1]

This articulation has never been chewed over for its precise meaning.  Rather, it has been swallowed whole by case law.  When we revisit it and break it down into its analytical elements, we see that it is a tautology laced with popular psychology.  Take a hard look at the above paragraph.  It reveals no more about bias than the sentence “I know it when I see it” educates the public of the crime of obscenity.[2]

The Supreme Court’s Delphic utterance has ensured the concept of bias has remained underdeveloped.  The cause? The weaving together of several arbitrary and perhaps ill-chosen concepts intended to sound good:

  • Reasonableness is a vague yet axiomatic standard the courts apply to decision-making. To say that bias or an attack on impartiality must be viewed by a reasonable and right-minded person simply casts a mirror unto its institutional self.  How else is a court to consider the issue – unreasonably?
  • More likely than not is an interesting importation of the balance of proof by which all civil and public law cases are determined, distinguishing it from the criminal or penal standard. It should be observed that the Supreme Court of Canada has recently updated the civil standard to require “clear, convincing and cogent” evidence to tip the scales in favour of the party bearing the burden of proof.[3]  Bias is a form of adjudicator misconduct, a breach of the duty to be impartial.  It is therefore a serious accusation to be leveled against a judge or arbitrator.  The courts have applied the “51%” standard without considering whether this or some other standard would be more appropriate.  (Eg., the “large majority” rule under Roberts Rules of Order for overruling a chair[4] or the enhanced civil standard employed by professional discipline tribunals.[5])
  • Whether consciously or unconsciously is a legal oddity because it assumes a type of mental state popularized by Freud which affects mood or behaviour in an unseen way. Perhaps this articulation of the legal test cracks open the floodgates to the introduction of expert psychoanalysis of judicial conduct.  Members of the judiciary are not, by training, capable of delving into a fellow judge’s unconscious.  Why put the unconscious in a legal test?
  • Would not decide fairly, the last and most operative words of the test, imports a negative consequence (not deciding fairly) triggered by the contaminant of the impartial state. Fairness stands alone as the justice value in peril.  Decision-making in other aspects of human endeavour, such as refereeing in sport, is held not only to a fairness standard but also one of correctness.  Is the capacity of bias to produce an incorrect result not as important to this type of decision-making as fairness?

The Legal Unconscious

The most peculiar and perhaps most influential aspect of the test is the role of the unconscious.  The reference to the unconscious mind originates in a Federal Court of Appeal formulation of the test for bias.[6]  The case involved the effect of prior corporate involvement in an energy consortium on the impartiality of a panel member of the National Energy Board.  Both the Court of Appeal and the Supreme Court appear to have taken as given that the workings of the unconscious are capable of impairing the ability of the conscious mind to disabuse itself of preconceptions formed during prior or external activities.

This understanding of the unconscious is consistent with a controversial psychological theory popular in the 1970s which held that mental processing of subliminal information plays a part in conscious judgments and behaviour.  This conscious-centric view of the unconscious posits a mind working as a traffic-controller of both objective and subjective information.  Hence, advertisers were rumoured to have implanted erotic symbols in depictions of ice cubes in drinking glasses, believing this would encourage sales of beverages.  The idea is that hidden codes can tap into the unconscious mind and translate into conscious decisions at the supermarket or liquor store.

This theory displaced a prior belief, dating back to the 1800’s, that the unconscious was a separate and unknown mental process that led to unintended behaviour and ideation, either through “Freudian slips” or hypnosis.  Any manifestation of the unconscious was seen as a form of illness, such as hysteria.  The 1970’s “conscious subliminal” view has, in recent years, been deemed too rigid, with research indicating that the unconscious can operate independent of the conscious mind.[7]  There is automaticity to this description of the unconscious mind, which is vaster in scope than the conscious focus on the mental object or task in hand.

The purpose of historically generalizing these “trends” in psychology is to illustrate that what the courts have considered to be unconscious influences on neutrality between two sides of a dispute are, in fact, part of consciousness and the cognitive ability to triage appropriate and inappropriate information.  If jurists were truly to consider the role of the unconscious on conscious impartiality, they would be delving into an area which the experts found unsatisfactory.  From the practical perspective of curial review of a complaint of bias, the role of the unconscious should best be abandoned, and each of the other constituent elements given greater scrutiny.

One can imagine the reference to the unconscious is meant as an indirect and intangible influence.  Even so, the word is meaningless in the context of law.  The unconscious is part of an attempt to dumb down the inquiry to give effect to popular misconceptions about the judiciary.  As part of the legal platform of the judicial system, the test offers no objective and practical guidance.  What is reasonable in an inchoate dispute is entirely in the eye of the beholder.

Recent Attempts to Triage Different Types of Bias

In the face of this need for a less subjective test, two recent court decisions have adopted a more objective and practical approach to the issue of bias.  The Ontario Superior Court held in Robinson v. Lepage that a deputy judge of the Small Claims Court had created an apprehension of bias, after a paralegal sponsored him in a charity cycling event and later appeared before him in court.[8]  The case was an inevitable consequence of a provincial government decision to deputize practising lawyers as part-time judges for Small Claims Court.  It was only a matter of time for these lawyers, who take on various community interests as part of a legal practice, to run into this type of problem when sitting as judicial officers.

In contrast, the Supreme Court of Canada ruled in Yukon Francophone School Board v. Yukon (Attorney General) that a trial judge’s participation as governor of a philanthropic francophone-rights group did not give rise to such an apprehension in a suit by a French-language school board for territorial underfunding of minority language schools.[9]  The court did give lip service to the 1976 “reasonable informed person” test.  However, the court decided the point on the basis that there was no evidence that the group had a specific policy aligned with one side of a dispute.

These recent cases get closer to defining the nature of bias in a manner that the law can apply practically.  The Yukon case did not concern bias as a result of monetary inducement.  Rather, it dealt with issue-directed bias.  A judge with active current involvement in a francophone linguistic rights entity was found not biased in a case involving minority French-language rights.  The case instructs us that the inquiry must not be a vague study in mind-reading or psychoanalysis.

Rather, the question of an apprehension of bias must refer to the evidence from which the archetypal reasonable person would have grounds to suspect bias.  The evidence must establish a mental alignment between the judge’s outside interest and that of one of the parties (or alignment against that of one of the parties, in another case).  The traditional test for bias requires an examination of both the conscious and the unconscious mind, a task slightly distinguishable from crystal-ball gazing.  In contrast, the issue-alignment form of bias is verifiable.

Somewhere between the external source of bias and the internal one is the plight of the judicial officer about whom a complaint has been lodged during the course of a hearing.  Each case must be considered on its own merits.  However, the paradox of the bias conundrum is that impartiality and independence require resilience on the part of the judicial or neutral officer.  Preserving the integrity of the process not only means keeping a high level of vigilance and transparency but also not bending to unjustified attempts to impugn one’s integrity.

Consider the following treatment by the Ontario Court of Appeal of lower court judges who pre-emptively recused themselves from presiding over a matter in which a self-represented litigant had complained to the Canadian Judicial Council about their ability to conduct a French-language hearing.  This was a particularly troublesome instance where the litigant had launched complaints about every available bilingual judge in the Toronto judicial region:

[13] In his reasons, the motion judge referred to the difficulties many judges encountered in cases involving the appellant and the fact that the appellant had complained about their French language skills. Although the motion judge made these comments, there is nothing in the record that would allow us to assess these assertions. Specifically, there is nothing in the record to indicate what happened before the motion judge or to verify the complaints that the appellant has apparently filed.

[14] In my view, the fact that the appellant has complained about the French language skills of several Toronto judges does not necessarily lead to a conclusion that all the bilingual judges in the region are precluded from hearing cases involving him. In fact, the appellant even criticized the motion judge who nevertheless [page499] proceeded to hear the motions. What’s more, the appellant has criticized judges of this court and still these judges continue to hear his appeals.

[15] The motion judge’s comments might be read as suggesting that the appellant’s criticisms amounted to an allegation of bias. However, criticism of a judge or an allegation of bias should not necessarily lead to the conclusion that a judge cannot preside, especially when this would lead to the loss of the right to a bilingual proceeding. Instead, a judge faced with such an allegation should receive supporting evidence and then decide, on a case by case basis, whether to recuse himself or herself.[10]

A More Evidence-Based Approach to Bias

What emerges from recent efforts to search for a principled basis for identifying judicial bias is a need for categorizing bias into different sources of unwanted influence in the judge’s role.  The traditional formulation of the concept has become too vague, too general and too subjective to be satisfactory.  A more searching analysis and formulation are beyond the scope of this article.  Nevertheless, it is possible to envisage, with more study, a more nuanced and more scientific approach to this important ethical principle of our legal system.


Adapted from a portion of Akazaki, “Overcoming Bias: Mediation-Arbitration in Canadian Civil Litigation,” published on SSRN (2015).


[1] Committee for Justice v. The National Energy Board, 1976 CanLII 2 (SCC), [1978] 1 S.C.R. 369 at 394

[2] Jacobellis v. Ohio, 378 U.S. 184

[3] F.H. v. McDougall, [2008] 3 SCR 41, 2008 SCC 53 (CanLII) at para. 46

[4] Roberts Rules of Order, Art. X, r. 58, online :

[5] Law Society of Upper Canada v. Stafford, 1999 CanLII 6899 (ON LST)

[6] Committee for Justice v. The National Energy Board, [1976] 2 F.C. 20, 65 D.L.R. (3d) 660 (F.C.A.), at paras. 20-22

[7] John A. Bargh and Ezequiel Morsella, “The Unconscious Mind,” (2008) Perspect Psychol. Sci. 2008 January; 3(1): 73-79, online

[8] Robinson v Lepage, 2015 ONSC 3128 (CanLII)

[9] Yukon Francophone School Board, Education Area #23 v. Yukon (Attorney General), 2015 SCC 25 (CanLII)

[10] Belende v. Patel (2008), 89 OR (3d) 494 (English) and 502 (French); 290 DLR (4th) 490, at 13-15



Terms of use / Mentions légales

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s