Rule of law in the Administrative Justice System – an ignis fatuus?

This post is the promised sequel to my May 28, 2014 post on the Federal Court decision in Muhammad v. Canada (Citizenship and Immigration), 2014 FC 448 (CanLII), concerning the law of independence and bias in the administrative justice system.  

The subject is what I characterized in the first post as the “doubtful reasons” the Federal Court (Strickland J.) offered for finding in Muhammad that the adjudicator in that case was independent and that there were no grounds for reasonably apprehending that the adjudicator was consciously or unconsciously biased.

I apologize for the unexpected delay.  It turned out to be a much longer project than I had anticipated.

In the following, regrettably very lengthy – but very interesting – analysis, I conclude that if the law of judicial independence and impartiality as the Court applied it in this judgment were to be accepted as the norm, we would have to concede that the rule of law in Canada’s administrative justice system is now no more than an ignis fatuus.  

To discover what an ignis fatuus rule of law looks like, read on.  (I have supplied numerous marginal headings to make the skimming easier.)


Risk of Torture

The case involves a judicial review of the decision of a Minister’s Delegate in the Department of Citizenship and Immigration (the Department) that Mr. Muhammad could be removed to Pakistan without risk of torture, a risk to life, or a risk of cruel and unusual treatment or punishment (all hereafter lumped together for convenience as “a risk of torture”).

The Adjudicator

The “Minister’s Delegate” was the final decider in the Department’s system for adjudicating applications for a Pre-Removal Review Assessment (PRR Assessment) – the last-ditch applications from individuals on the verge of being removed from Canada who claim they face a risk of torture if they are sent home.

Risk of Torture a Justiciable Factual Issue – Not a question of Ministerial Discretion

It is common ground that notwithstanding that the adjudicator in this case is referred to in the judgment as the “Minister’s delegate”, she was not exercising a Ministerial discretion but adjudicating a justiciable factual issue. The issue was whether, on the evidence, on a balance of probabilities, there was a risk that Mr. Muhammad would be tortured if he were removed to Pakistan.  And, of course, whichever way that decision went on that issue it would be a life-altering decision.

(To keep the administrative justice focus sharp, I shall hereafter principally refer to the “Minister’s delegate” as the “adjudicator”.)

An Independent and Impartial Adjudicator is Required

In these circumstances, the court acknowledged that the principles of procedural fairness required that the adjudicator be independent and impartial.  Thus, given the established law on these matters, the issues for the court were:

1)   The Independence Issue:

Could the adjudicator’s independence be reasonably seen to be protected by sufficient objective guarantees of the three essential conditions of independence defined by the SCC in Valente – security of tenure, financial security, and administrative or institutional independence?

Or, more simply, and to paraphrase the Supreme Court of Canada’s statement of the issue in Généreux (of which much more later):

 Under the structure of the [PRR Assessment] system … would a reasonable person have been satisfied that the [Minister’s Delegate/adjudicator] was independent [from the executive branch]?

(2)   The Bias Issue:

If the adjudicator’s status satisfies the conditions of independence, were there grounds for reasonably apprehending that the adjudicator would not be impartial – that is, that she would be consciously or unconsciously biased – notwithstanding her independence?


The Wanted List Gets its Man

Mr. Muhammad had been living illegally in Canada and in 2011 he was discovered and arrested after his name appeared on the government’s recently conceived “Wanted List”.

Deportation Resisted Because of the Risk of Torture

Being scheduled to be returned to Pakistan, Muhammad applied for a PRR Assessment arguing that under Canada’s immigration laws he could not be removed to Pakistan because if he were, he would be at risk of torture, etc.  His reading of the law was correct: if the PRR Assessment process led to a decision that on a balance of probabilities there was a risk he would be tortured in Pakistan, he would it seems be entitled to stay in Canada.

Not a Terrorist

Muhammad’s right to stay in those circumstances was subject to the further condition that there be no information that links him directly to any terrorist organization’s crimes against humanity or terrorist acts or that was sufficient to establish that he was a danger to the security of Canada.  However, it is acknowledged that in Mr. Muhammad’s case that condition has been met.

Big Surprise for the Government

To the government’s obvious consternation, the Department’s PRR Assessment Officer who assessed the torture risk concluded there was a risk of torture becauseand only because – the applicant had been placed on the government’s Wanted List.

The Minister’s Delegate Makes the Final Decision

But the PRR Assessment Officer’s decision was not conclusive – it appears to have the status essentially of an opinion.  The final decision on the torture issue falls within the jurisdiction of the “Minister’s Delegates” – or as I have chosen to call them the “adjudicators” – after considering the evidence afresh.

This is the Second Decision for Muhammad

The adjudicative decision under review by the Federal Court in Muhammad is in fact a redetermination of an earlier adjudicative decision on the same issue.  In the original decision a different Minister’s Delegate/adjudicator had also found there was no risk of torture.

On judicial review of the latter decision the federal court had found the decision unreasonable for lack of sufficient justification and sent it back for a redetermination by a different adjudicator.

The argument

In this second of the two judicial review proceedings, the applicant argued that the redetermination decision was invalid because the adjudicator was not independent and could not be seen to be impartial.  He also argued, in the alternative, that, like the first adjudicator’s decision, this decision was also unreasonable.

The Outcome

As I have said, the court in this instance  ruled that the adjudicator was independent and that there were no reasonable grounds for apprehending bias.  However, it also concluded that the redetermination decision was unreasonable – again for lack of adequate justification.  The court sent the issue back for a second redetermination by a third, different adjudicator.

It is, of course, the ruling on the independence and bias issues that is the subject of this post.


Remember the Salient Point

In reviewing these facts, remember the salient point: it is not disputed that we are talking here of an adjudication of a factual issue of life-altering importance to which the rule of law’s requirement for independent adjudicators indisputably applies.

The Facts

The facts, as found by the Court, that are relevant to the independence issue include these:

The Minister – the Executive Branch – is the Opposing Party

  •  There were two parties to the proceedings presided over by this adjudicator – the applicant, Muhammad, and, opposing him, the respondent, the Minister of Citizenship and Immigration Canada (the Minister).

The Minister’s participation in the Hearing is Managed by his Department’s Case Management Branch

  • The management of the Minister’s litigation in high profile cases is the responsibility of the Department of Citizenship and Immigration’s Case Management Branch.  That responsibility includes managing the litigation of PRR Assessment applications.

The Case Management Branch (CMB) Employs a Cadre of Adjudicators

  • The latter responsibility also included supplying the adjudicators for those applications.  The PRR Assessment adjudicators come from a group of Case Management Branch employees appointed by the Branch to the position of “Director, Case Determination”.
  • There are a number of such “Directors” and the respondent Minister has delegated the power to adjudicate PRR Assessment applications to those Directors which explains why the adjudicator in this case is referred to in the judgment as the “Minister’s Delegate”.

The CMB Head Supervises the Adjudicators and Evaluates Their Performance

  • The PRR Assessment adjudicators in the Case Management Branch are supervised directly by the head of the Case Management Branch. They report to him personally and meet with him on a “regular” basis to “discuss operational matters and individual files”.  It is also the Branch head who conducts their “mid-year and year-end performance reviews”.

The CMB Head Picks the Adjudicator for Each Case

  • There is no independent structure for selecting the particular adjudicator to be assigned to a particular case.  The adjudicators assigned to Muhammad’s case were the arbitrary choices by the adjudicator’s supervisor – the head of the Case Management Branch.

Hallmarks of Dependency

  • In an affidavit filed in this case, the head of the Branch stated that he had “advised the Minister’s Delegates (the Directors of Case Determination in his Branch – the “adjudicators”, as I call them) not to discuss their cases with him; that he has never discussed the contents of their decisions; and that his practice is to emphasize to the [adjudicators] that their decisions are theirs alone”.


The Relevant Essentials

The facts in Muhammad as set out above but reduced to the essentials relevant to the issue of whether a reasonable person would be satisfied that the adjudicator was independent of the executive branch, may, in my submission, be fairly described as follows.

While the executive branch was the opposing party in the PRR Assessment proceedings it nevertheless got to choose the “judge”. The judge, who was a member of the executive branch*, was chosen and assigned to the adjudicative position by her executive branch supervisor.  This was same supervisor who was responsible for the management of the executive branch’s opposition interests in the proceedings and whose duties included evaluating the adjudicator’s performance.

*Although, it must be noted that in her judgment, in para. 144, Strickland J. referred to and followed Federal Court jurisprudence that has apparently held that members of the Public Service of Canada are not part of the executive branch…  See the discussion below.

Généreux – The Missing but Nonetheless Dispositive Authority

The law of judicial independence as it applies to these facts is most usefully seen in a 1992 Supreme Court of Canada decision in which the Valente principles of judicial independence were applied to a set of facts that are, in all relevant respects, indistinguishable from the facts in Muhammad.

The decision is R. v. Généreux, [1992] 1 SCR 259.   It is the decision in which the Supreme Court found the traditional structures of the Canadian Army’s General Court Martial proceedings to contravene the judicial independence requirements spelled out in Valente.

The decision overruled the Court’s own pre-Valente decision in MacKay and led to a major restructuring of what had been a long tradition of executive-branch domination of military justice.

Généreux was not referenced but it is plainly dispositive of the independence issue in Muhammad.

The Executive Branch’s Roles in the Généreux Adjudication

In Généreux, the “applicant” was a soldier charged with a narcotics offence.  The body opposing the applicant – i.e., the body that convened the court martial and prosecuted the applicant – was, the Court found, an “integral part of the military hierarchy.  And because the senior members of the military hierarchy were, the Court also found, “responsible to their superiors in the Department of Defence” the hierarchy was therefore, in the Court’s view, “an integral part of the executive”.

The management of the court martial proceeding was in the hands of the Office of the Judge Advocate General, whose “close ties” to the “executive” were, in the Court’s view, “obvious”.

The “judge” in the case was the person designated as the “judge advocate”.   He was a military judge, which is to say, a military “legal officer” employed within the Office of the Judge Advocate General and who had been designated by the Judge Advocate General to be a member of the Office’s cadre of “military judges”.

The selection of individual military judges to be the judge advocate in particular cases was, the Court found, the responsibility of the “executive” – i.e., the Judge Advocate General.

Moreover, a military judge’s performance was formally evaluated by his or her superiors, also part of the executive – an evaluation that “could”, the Court said, “potentially reflect his superior’s satisfaction or dissatisfaction with his  conduct at a court martial” with the result that “the executive might effectively reward or punish an officer for his or her performance [as the judge advocate] of a General Court Martial”.

The Généreux facts are the Muhammad Facts – a Perfect Fit

A comparative analysis of the two cases will show that the role played in Généreux by the Minister of Defence was played in Muhammad by the Minister of Citizenship and Immigration; the role of the military hierarchy in Généreux was played in Muhammad by the Department of Citizenship and Immigration; the role of the Office of the Judge Advocate General was played by the Case Management Branch, the role of the Judge Advocate General by the head of the Case Management Branch, and the role of the judge advocate – the judge – was played by the Minister’s Delegate – the adjudicator – chosen by the head of the Case Management Branch from the Branch’s cadre of adjudicators.

In short, in both cases the applicant’s interests were opposed by the executive branch and adjudicated by the executive branch.

The Généreux Result

On these facts, the Supreme Court found in Généreux that none of the three Valente conditions of judicial independence – security of tenure, financial security, or institutional independence – could be seen to be satisfied.

An exposition of the law of judicial independence as the Généreux Court applied it to those facts may be found in the majority judgment of Chief Justice Lamer.

Of special relevance to the Muhammad facts is the following expression in the Chief Justice’s judgment in Généreux of the Court’s concern about the executive being free to choose arbitrarily the particular adjudicator to be assigned to a particular case.

 … As a result, there was no objective guarantee that [the judge advocate’s] career as military judge would not be affected by decisions tending in favour of an accused rather than the prosecution.  A reasonable person might well have entertained an apprehension that a legal officer’s occupation as a military judge would be affected by his or her performance in earlier cases.  Nothing in what I have said here should be taken to impugn the integrity of the judge advocate who presided at the appellant’s trial, nor to suggest that judge advocates in fact are influenced by career concerns in the discharge of their adjudicative duties.  The point is, however, that a reasonable person could well have entertained the apprehension that the person chosen as judge advocate had been selected because he or she had satisfied the interests of the executive, or at least has not seriously disappointed the executive’s expectations, in previous proceedings.  Any system of military tribunals which does not banish such apprehensions will be defective in terms of [the requirements of judicial independence … (Emphasis Added.)


The facts in Muhammad as found by the Court that are relevant to the bias issue include the following.

The Minister’s Stake in the Wanted List

It was well known within the Case Management Branch that if the Branch adjudicators were to agree with the PRR Assessment Officer’s decision that Mr. Muhammad was not removable from Canada because he had been placed on the Wanted List, the Minister of their department would suffer a significant political embarrassment.  This was the department in which the adjudicators (and their supervisor, the head of the Branch) are employed and in which they are all pursuing their public service careers.

The Minister had publicly defended the Wanted List against the very criticism that the PRR Assessment Officer’s decision in Muhammad now showed to have been prescient – that is, that listing a person on the Wanted List would open him to a risk of torture and thus prevent his removal from Canada.

What The Adjudicator Knew

At the time she made her decision, the adjudicator knew of the Minister’s public support of the Wanted List and of his public dismissal of the concern that listing a person on the List would create a risk of torture and effectively prevent that person from being removed from Canada.

The CBSA Takes Action – The “Ill-Advised” Meeting

The Wanted List is an enforcement tool devised and administered by the Canada Border Services Agency (CBSA).

After the PRR Assessment Officer’s decision in Muhammad was published, the “Director General of the CBSA’s Border Operations” sought what is acknowledged to have been an unprecedented and, in the court’s words, an “ill-advised” meeting with the head of the Case Management Branch.

The Whistleblower’s Intervention

The applicant’s counsel became aware of this meeting through an affidavit sent to him by a former director of enforcement with the CBSA in what one assumes was an act of whistleblowing. (This information came to light three weeks after the redetermination decision had been made and in time to be included as grounds for the bias claim in the judicial review application.)

The Court’s View

The Court’s own criticism of the meeting reads as follows (para. 153):

The February 3, 2012 meeting was certainly ill-advised as it could easily be perceived as, and indeed may have been, an attempt to influence the decision-making process.

The Evidence

The evidence of that meeting consisted of the whistleblower’s affidavit and the cross-examinations of the affidavits subsequently filed in response by all those who had been at the meeting.  That evidence is recounted by the Court in paras 122 and 123 of the judgment.

The stated agenda for the meeting was: “Muhammad – discussion on next steps”.  (The most important “next step” at the time would have been the final adjudication of the risk issue by a Case Management Branch adjudicator.)

What Happened at the Meeting?

At the meeting, the CBSA’s Director-General of Border Operations commented negatively on the Case Management Branch’s lack of oversight of the PRR Assessment Officer’s decision and expressed her concern about the impact on the CBSA’s enforcement mandate should the Officer’s decision be confirmed by a Case Management Branch adjudicator thereby putting the viability of the Wanted List in question.

The CBSA Comes Away from the Meeting Sure of a “Good Decision”

One of the senior members of the CBSA (the head of the CBSA’s Case Management Division) who had accompanied her boss to the meeting testified that she left the meeting confident that there would be “a good decision” for the CBSA in the Muhammad case.  (Para. 153)

But no Evidence that Adjudicator Knew of the Meeting

The Court found, however, that there was “no evidence” that the concerns expressed to the head of CMB by the CBSA’s officials’ in their ill-advised meeting with him had been conveyed to the adjudicator.  (It is not without interest, however, that the Court does not say that either the CMB head or the adjudicator had testified to that effect.)

The Executive Branch Adjudicators Always find no Risk of Torture

The Court also had before it a Statutory Declaration filed in support of the application by a lawyer with the Applicant’s counsel’s firm stating that where PRR Assessments were performed by Case Management Branch adjudicators in the context of security certificate cases, his law firm’s experience was that those adjudicators “always”, and unreasonably, found that the applicants faced no risk of torture upon deportation:

Post Suresh, where either the PRRA assessment or the danger opinion was made in the context of Security Certificate Cases, the Minister’s Delegate always found that there was no risk of torture faced by the individual named in the Certificate.  All of the persons represented by my firm in this situation were subsequently successful in obtaining stays of their removals from the Federal Court due to the unreasonableness of the Minister’s Delegate’s finding.


Newfoundland Telephone

An authoritative statement of the bias principle and the test for bias as it applies to individual adjudicators may be conveniently found in Newfoundland Telephone Co. v. Newfoundland (Board of Commissioners of Public Utilities), [1992] 1 SCR 623 at page 636:

 The duty to act fairly includes the duty to provide procedural fairness to the parties.  That simply cannot exist if an adjudicator is biased.  It is, of course, impossible to determine the precise state of mind of an adjudicator who has made an administrative board decision.  As a result, the courts have taken the position that an unbiased appearance is, in itself, an essential component of procedural fairness.  To ensure fairness the conduct of members of administrative tribunals has been measured against a standard of reasonable apprehension of bias.  The test is whether a reasonably informed bystander could reasonably perceive bias on the part of an adjudicator. (Emphasis added.)


The Supreme Court has addressed the law respecting the apprehension of bias most prominently and most recently in Wewaykum Indian Band v. Canada, [2003] 2 SCR 259.

You will recall that in that case the appellant Indian Bands had applied to have a decision of the Supreme Court set aside on the grounds of an apprehended bias on the part of Justice Binnie.  The bias concern arose because of Justice Binnie’s prior involvement in the Bands’ issue in the years when he was the Associate Deputy Minister of Justice – an involvement that Justice Binnie did not remember.

Why is it that the Law Requires Only an Apprehension of Bias?

The Court took the opportunity occasioned by its consideration of the allegations against Justice Binnie to examine why, for the law to disqualify an adjudicator on the grounds of bias, it does not require proof of actual bias but only proof that there is a reasonable apprehension of bias.

It performed this analysis through the lens provided by the question of what would motivate applicants to say that they are not alleging actual bias but are just claiming an apprehension of actual bias.  And it saw three reasons that explained applicants choosing not to assert actual bias. (See Wewaykum para. 62.)

 One, it is Unwise and Unrealistic to Require Proof of Actual Bias

The first reason for not attempting to prove actual bias arises when the circumstances are such that it would be “unwise or unrealistic” for the law to require “proof of the real thing”.  In those circumstances, the Court recognizes that “reasonable apprehension of bias” is a “surrogate” for actual bias where actual bias is not reasonably or appropriately provable. (See Wewaykum, para. 63.)

 Two, Unconscious  Bias Disqualifies an Adjudicator but Cannot be Proved

The second reason for applicants relying only on an apprehension of bias arises, the Court said, in the circumstances where the applicant may be prepared to concede that the judge acted in good faith but was nevertheless unconsciously biased. There is obviously no chance at all, the Court acknowledged, of proving an unconscious bias, but a judge who has one is nevertheless disqualified, and the only way to get at an unconscious bias is through proving a reasonable apprehension of bias.

The latter point was addressed directly in Wewaykum in the following passage (Para. 65):

 … In R. v. Gough, [1993] A.C. 646 (H.L.), at p. 665, quoting Devlin L.J. in The Queen v. Barnsley Licensing Justices, [1960] 2 Q.B. 167 (C.A.), Lord Goff reminded us that:

 Bias is or may be an unconscious thing and a man may honestly say that he was not actually biased and did not allow his interest to affect his mind, although, nevertheless, he may have allowed it unconsciously to do so.  The matter must be determined upon the probabilities to be inferred from the circumstances in which the justices sit.

 And Three, the Court Recognizes that Real Bias is not the Relevant Inquiry

The third reason the Court identified for an applicant claiming only an apprehension of bias is the recognition that looking for real bias is “not the relevant inquiry”.  See the Wewaykum Court’s summary of that position in the following passage: (para. 66.)

 In the present case, as is most common, parties have relied on Lord Hewart C.J.’s aphorism that “it is not merely of some importance but is of fundamental importance that justice should not only be done, but should manifestly and undoubtedly be seen to be done” (The King v. Sussex Justices, Ex parte McCarthy, [1924] 1 K.B. 256, at p. 259).  To put it differently, in cases where disqualification is argued, the relevant inquiry is not whether there was in fact either conscious or unconscious bias on the part of the judge, but whether a reasonable person properly informed would apprehend that there was.  In that sense, the reasonable apprehension of bias is not just a surrogate for unavailable evidence, or an evidentiary device to establish the likelihood of unconscious bias, but the manifestation of a broader preoccupation about the image of justice.  As was said by Lord Goff in Gough, supra, at p. 659, “there is an overriding public interest that there should be confidence in the integrity of the administration of justice”.

Bias and the Standard of Proof

In Wewaykum, the Supreme Court also addressed the standard of proof issue in apprehension-of-bias cases, viz:

 76   …it is worth repeating that the standard refers to an apprehension of bias that rests on serious grounds, in light of the strong presumption of judicial impartiality.  In this respect, de Grandpré J. added these words to the now classical expression of the reasonable apprehension standard:

 The grounds for this apprehension must, however, be substantial, and I . . . refus[e] to accept the suggestion that the test be related to the “very sensitive or scrupulous conscience”.

 Not Very Sensitive but not Insensitive Either

The author (Ellis) would note that the ruling out of bias apprehensions of an objective observer because they are seen to be attributable to an overly sensitive or overly scrupulous conscience does not justify substituting the apprehensions of an objective observer whose conscience is insensitive to the dangers of adjudicator bias.

Presumably what is wanted is the classic reasonable person – one who is not overly sensitive but not unduly insensitive either.

The Degree of an Observer’s Sensitivity to Bias Must Adjust to What is at Stake

Moreover, the author would argue, we must expect the sensitivity of any objective observer to the danger of bias to be reasonably affected by his or her understanding of what is at stake.  And it stands to reason that such an observer`s sensitivity will be enhanced in the circumstance where the bias concerns are known to relate to an adjudicator entrusted with an adjudication in which a conscious or unconscious bias in favour of the respondent could result in the applicant being exposed to life-altering consequences such as, in this case, a risk of torture.


The Third Rail of the Independence Issue

Of the three conditions of judicial independence defined in Valente, it is perfectly apparent that the adjudicator in Muhammad could not satisfy the third one; and she only needed to fail one to be disqualified.

(Moreover, it is more than probable, though perhaps arguably not conclusive, that she could not satisfy the first two either.  On its equivalent facts, the Supreme Court held in Généreux that the judge advocate had not satisfied either of the first or second conditions.)

The third condition which the adjudicator plainly could not satisfy is what Valente called variously “institutional” independence or “administrative” independence or “adjudicative” independence.  The requirement is defined in Valente, paragraphs 47 to 52.  Its traditional expression is to be found in paragraph 52, viz:

 The third essential condition of judicial independence … is in my opinion the institutional independence of the tribunal with respect to matters of administration bearing directly on the exercise of its judicial function.

And what the Court had in mind when it referred to “matters of administration bearing directly on the exercise of its judicial function” clearly included the assignment of adjudicators to particular cases. Para. 49 reads in part as follows:

Judicial control over the matters referred to by Howland C.J.O. [in Valente (No. 2)] ‑‑  assignment of judges, sittings of the court, and court lists ‑‑  as well as the related matters of allocation of court rooms and direction of the administrative staff engaged in carrying out these functions, has generally been considered the essential or minimum requirement for institutional … independence. (Emphasis added.)

In Muhammad, the Third Rail was Emphatically Touched

Of course, in Muhammad, the “assignment” of the “judge” was not under the control of an independent body but directly under the control of the executive branch itself.

The law as it appears in the passage quoted earlier from Généreux re the assignment of the judge advocate to a particular case by the “executive” applies at least equally to the selection of the adjudicator in Muhammad by the head of the Case Management Branch, viz:

 …  The point is … that a reasonable person could well have entertained the apprehension that the person chosen as [the adjudicator] had been selected because he or she had satisfied the interests of the executive, or at least has not seriously disappointed the executive’s expectations, in previous proceedings.  Any system [of adjudication]… which does not banish such apprehensions will be defective in terms of [the requirements of judicial independence]….

It is also important to remember in this connection that the person who selected the adjudicator to redetermine the torture issue is the person – the head of the CMB – who had been at the ill-advised meeting with the CBSA officials; a meeting that had left one of the CBSA officials with the impression (presumably derived from the CMB head himself since he was the only CMB representative at the meeting) that the CBSA could count on getting a “good decision” from the CMB adjudicators on the torture issue.

And, Indeed, is it not Painfully Obvious?

Moreover, if one looks at the independence issue more generally – viewed say from 1,000 feet rather than the 500 foot altitude at which one might say the three specific conditions of independence come into focus – is it not just painfully obvious that this adjudicator was not independent?  Consider:

  • That the adjudicator was selected to be the adjudicator by the opposing party in the proceedings – that is, the government;
  • That she is in fact an employee of that opposing party;
  • That there is at least a reasonable apprehension that the government’s satisfaction with her performance in adjudicating the same or similar issues in the past is likely to have influenced her selection to do so in this case;
  • That her adjudicative career requires her to get further appointments from the opposing party in future similar cases; and
  • That an objective observer would reasonably perceive that the adjudicator would believe her career prospects to be in some degree contingent on the continuing approval of the opposing party (as effectuated in particular through the vehicle of the Branch head’s biannual evaluation of her performance).

The Judgment on the Independence Issue

Strickland J.’s analysis of the independence issue appears in her judgment under the marginal heading: “Structural Independence or Impartiality, Individual Independence” and is to be found in paragraphs 124 to 144 of the judgment.

Her conclusion, which appears in para. 124, is that “the Applicant’s argument that there is a lack of structural independence … as a result of situating the Minister’s Delegate in the CMB office cannot succeed”.

It is at this point that the rule of law begins to take on the hue of an ignis fatuus.

From a doctrinal perspective, the problem is that Strickland J.’s analysis and the analysis of her colleagues’ decisions in the federal courts in similar cases on which she relies do not deal at all with the Valente requirement of institutional independence, but only with the law relating to institutional impartiality.

This is the law originating in the SCC decision in Lippé where the concept of institutional bias was first recognized and where it was held that to establish institutional bias one must show that there will be a reasonable apprehension of bias in a substantial number of cases.

While Justice Strickland’s judgment refers to “structural independence and impartiality”, it in fact conflates the two issues into only the one: the institutional impartiality or, to put it the other way round, “institutional bias” issue.

Paragraph by Paragraph Account of the Judgment on the Independence Issue.

Para. 126:

The paragraph cites Justice Cory in Rv.S(RD):

… the threshold for a finding of real or perceived bias is high … the reasonable person must be an informed person with knowledge of all the relevant circumstances, including the traditions of integrity and impartiality that form a part of the background and would be apprised also of the fact that impartiality is one of the duties the judges swear to uphold.

        Author’s Comment:

In RvS(RD), the Supreme Court was dealing with an allegation of actual bias against a provincial court judge.

In the independence jurisprudence – Valente and Généreux, etc – there is no mention of a high standard of proof.  Indeed it could be argued that, once an applicant has identified a prima facie appearance of dependency in an adjudicator, the rule of law should be seen as requiring the government to then satisfy the court that that perception is somehow misguided.

And the major point in Valente is that the courts’ traditional reliance on tradition and duty as grounds for presuming judicial independence is not enough:

“while tradition reinforced by public opinion may operate as a restraint upon the exercise of power in a manner that interferes with judicial independence, it cannot supply essential conditions of independence for which specific provision of law is necessary”. (Valente, para 36.)

(And see more generally Unjust by Design at page 66-71.)

It is also worth noting that when the adjudicator who is alleged to be biased is a provincial court judge, it is reasonable for a court to factor in – take judicial notice of – the tradition of independence and integrity associated with such judges.  But where is the justification for taking judicial notice of a tradition of independence in the case of the Minister’s Delegates?

It is also not without significance that there is no suggestion in Muhammad that the PRR Assessment adjudicators have sworn to be impartial. The evidence is that the Muhammad adjudicator was told in her training the importance of impartiality; knows that she should be impartial and appear to be impartial; and claims to have acted impartially; but not that she has sworn to be impartial.

Para. 127:

This paragraph, citing Justice Gibson in Say, makes the following points: the standard is a reasonable apprehension of bias not viewed through the eyes of a person of very sensitive or scrupulous conscience; grounds must be substantial; and “substantial deference is owed to Government decisions that relate to appropriate organization of public servants devoted to the administration of the vast range of responsibilities of the Government of Canada.(Emphasis added.)

          Author’s comment:

Accepting that a government’s choice of structure for independent administrative justice adjudicators is to be shown the same “substantial deference” as for the government`s organization of its other vast range of responsibilities, and reviewed only through the eyes of persons not overly sensitive to rule of law issues,  constitutes, I would suggest, a significant contribution to rendering the rule of law in the administrative justice system an ignis fatuus.  

Para. 128:

This paragraph quotes with approval Gibson J. in Say: “… the only evidence adduced … tending to support institutional bias or want of impartiality and independence was anecdotal at best; contrast the evidence that adjudicator training included the importance of impartiality and independence. (Emphasis added.)

           Author’s comment:

Here is the first indication of Strickland J.’s error in seeing evidence that the adjudicator believed herself to be independent and not biased as relevant to the independence issues.  Since Valente, this is clearly not the issue.  An adjudicator cannot decide to be independent, and the fact that an adjudicator knows that he or she is supposed to be independent is immaterial; it is a question of his or her status – of whether his or her structural relationships with the government provide objective guarantees of independence.

To quote Généreux :

 … the question is not whether the [tribunal] …actually acted in a manner that may be characterized as independent and impartial.  The appropriate question is whether the tribunal, from the objective standpoint of a reasonable and informed person, will be perceived as enjoying the essential conditions of independence.

 …   The question raised in this appeal, however, is not resolved by this earlier case [SCC in MacKay holding that court martial structures were independent.*]   First, the majority of this Court in MacKay seems to have applied a subjective test.  It asked whether the Standing Court Martial actually acted in an independent and impartial manner.  This is not, in light of Valente, the appropriate test.  …

*MacKay was overruled by the Court in Généreux on the basis of Valente having changed the law.

Para. 129:

Still citing Gibson in Say: the structural independence and impartiality issue is whether “a fully informed person would have a reasonable apprehension that bias would infect decision makers in the PRRA program in a substantial number of cases”.

Para. 130:

Quoting Blanchard J. in Singh and thereby again showing that Justice Strickland sees the institutional issue to be solely one of institutional bias – “an allegation of bias is of such momentous importance … the grounds … must [therefore] be substantial. “

          Author’s Comment:

 Should it not be the other way around – that is, that an allegation of a structural dependency of an adjudicator on the government is of such “momentous importance” that in the face of a prima facie case, a Court should require the government to provide substantial proof that the dependency is not material?

Para. 134:

“… given that an allegation of a lack of institutional impartiality is of such potential significance from both an operational and a procedural fairness perspective, the grounds to establish it must be substantial. The evidence adduced by the Applicant in this case is insufficient to meet this requirement and satisfy his onus of demonstrating want of impartiality in a substantial number of cases. The mere fact that the Minister’s Delegate is situated in the CMB, particularly when considered together with the evidence concerning her relationship to and communications with both [the head of the Case Management Branch] and the Minister’s Office, does not meet the onus.”  (Emphasis added.)

          Author’s comment:

 For the evidence concerning the adjudicator’s relationship to and communications with the head of the Case Management Branch see above.

 Again, Strickland J. is not addressing the issue of institutional independence.

 Para. 135:

In this paragraph Justice Strickland dismisses the affidavit evidence that the PRR Assessment’s Minister’s Delegates/adjudicators “always” find no risk of torture.  The decisions referenced have not, she says, been identified; the bases for those decisions were not provided; each decision is based on the individual facts…  In short, “[This evidence] is insufficient to establish a want of independence in a substantial number of cases”.

 Para. 137:

Where a substantial number of cases cannot be identified, allegations of an apprehension of bias cannot be brought on an institutional level, but must be dealt with on a case-by-case basis (Benitez v Canada (Minister of Citizenship and Immigration), 2006 FC 461 (CanLII), 2006 FC 461 at para 196; Lippé, above). Here, this involves a consideration of whether the Minister’s Delegate lacked the hallmarks of independence, those being security of tenure, financial security and administrative control (Matsqui, above, at para 73, 75), and whether there was a reasonable apprehension of bias or abuse of process as a result of interest in the wanted list. (Emphasis added.)

     Author’s Comment:

While in this paragraph, there is finally mention of the conditions of independence – security of tenure, financial security and “administrative control” – the ensuing six paragraphs go back to dealing with the bias issues presented by what Strickland J. refers to here as the “interest in the wanted list”, and which she examines through both the bias and the abuse of process lens.  She in fact deals with the issue of whether the adjudicator in this case “lacked the hallmarks of independence” only in Paragraph 144.  (See below.)

Para. 138:

This paragraph quotes the Justice de Grandpré test for a reasonable apprehension of bias.

 Para. 139:

Here Justice Strickland sets out the law relating to the abuse of process.

      Author’s Comment:

This post is not concerned with the abuse of process aspects of the ill-advised meeting.

 Para. 140:

“It is not disputed that there was considerable government interest in the CBSA’s wanted list and that there were concerns about the implications of a positive risk assessment on the list. It is therefore certainly not outside the realm of possibilities that, given this interest, a decision-maker could be inclined toward a certain result in the absence of sufficient hallmarks of independence.”

 Para. 141:

“However, the Applicant has not put forth any evidence to demonstrate that the Minister’s Delegate was not independent and impartial. Absent evidence to the contrary, a decision-maker is presumed to be impartial (Mugesera, above). Allegations of a lack of independence or a reasonable apprehension of bias are serious and cannot be based on pure speculation or limited evidence. Here, the Applicant’s submissions in this regard are also rebutted by the evidence of the Minister’s Delegate, [the head of the Case Management Branch] and others. (Emphasis added.)

     Author’s Comment:

 With respect, as far as the independence issue is concerned, this statement of the law is not correct.   Again one sees the reference to “independence and impartiality” but the issue of independence being once again conflated with impartiality.  After Valente, no adjudicator is presumed to be independent.  And after Valente and Généreux the independence question is not whether the adjudicator acts independently, or sees herself to be independent, or whether she is treated by the government as independent.

Independence is, as I have said and the Courts have made clear, a question of status – the structure of the tribunal’s relationships to the executive branch – and whether a reasonable person would be satisfied that those structures not only objectively guarantee the adjudicators who presided in that tribunal their personal independence from the executive branch – that is, guarantee their personal security of tenure and financial security – but also guarantees that the “tribunal” over which they preside has institutional control over the administrative functions directly relevant to the adjudication process.

 In particular, on these facts, the law asks: does the government control the choice of the adjudicator who will preside in that tribunal in a particular case?.

The answer to the latter question, on the Court’s own findings of fact, is clearly yes, and yet nowhere in the judgment does the Court address what is referred to in the Valente jurisprudence as the assignment-of-judges issue.  If it is the government who assigns the judges that preside over a tribunal’s particular case, the law is perfectly clear: that tribunal does not qualify as an independent tribunal.

 In this case, the adjudicator was the person who was assigned to preside in the PRR Assessment tribunal’s hearing of Muhammad’s risk issue, and nobody disputes that she was assigned to that role by the executive branch.  With respect, it should have been an open and shut case on that ground alone.

Para. 144:

“In Mohammad, above, the Federal Court of Appeal found that the adjudicator in that case, who was an immigration officer pursuant to the IRPA, had security of tenure, which is generally available to public servants. Similarly in Dunova, above, described in greater detail below, Chief Justice Crampton found that PRRA officers are independent as they are members of the Public Service of Canada which is independent from the executive branch of government. Here, the Minister’s Delegate is also a member of the Public Service of Canada and therefore, by corollary, the same principles apply.” (Emphasis added.)

(Note: the reference to “Mohammad” in the above quote is to an 1988 decision of the Federal Court of Appeal, not to any proceeding related to Mr. Muhammad in this case.)

      Author’s Comment:

With great respect, I just don’t know what to say.  To suggest that individuals who are members of the Public Service and employed in government departments are for that reason not part of the executive branch – are independent of the executive branch – is simply beyond understanding.

It is axiomatic that the “executive branch consists of the Prime Minister and Cabinet and the “administration”.  As far as I know, everyone in the “administration” is a member of the Public Service – presumably including in this case, both the adjudicator and the head of the Case Management Branch.

 Justice Strickland did not address the financial security requirement but presumably she would see it to be guaranteed as well by the adjudicator’s status as a member of the Public Service of Canada.

But to suggest that the adjudicator’s security of tenure and financial status are guaranteed by that status is to ignore the inevitable impact on her career as a public servant of the performance reviews performed twice a year by her superior – the head of the Case Management Branch.

 Justice Strickland does not address the implications of those biannual reviews by the person who is responsible for managing the litigation of the Minister’s high profile cases, but an objective observer would reasonably perceive that the adjudicator could well believe that a decision by her that was seriously at cross purposes to the head’s and his branch’s, and his (and her’s) Minister’s interest in the Muhammad case could have some significant career implications for her.

 In any event, as I have indicated above, the executive branch’s role in assigning the adjudicator to the case is enough to disqualify the adjudicator on independence grounds, whatever one might think of the guarantees of her security of tenure and financial security.


The law concerning reasonable apprehensions of bias is set out above and is not contentious.

 Adjudicators who give an objective, not unduly sensitive, observer reasonable grounds for apprehending that they will be likely to be consciously or unconsciously biased are disqualified as adjudicators – disqualified  because they fail to satisfy the procedural fairness principle that adjudicators be and be seen to be impartial.

Why Might One Think that the Muhammad Adjudicator was Consciously or Unconsciously Biased?

From the undisputed facts set out above, the grounds that might reasonably be grounds for apprehending bias may be briefly summarized as follows.

The adjudicator could reasonably be seen to have known that deciding in this case that there was a risk of torture caused by Muhammad being included on the Wanted List would politically embarrass her Minister in a serious way, make her boss mad and threaten his career, and possibly have a negative effect on her future career as an adjudicator in the Citizenship and Immigration Department.

The other circumstance that an objective observer might be concerned about in considering the issue of bias is, of course, the so-called “ill-advised” meeting between the CBSA officials and the head of the Case Management Branch following the publication of the PRR Assessment Officer`s original decision.

The Court’s Reasons on the Bias Issue

Justice Strickland’s reasons for finding that there was not reasonable ground for apprehending a conscious or unconscious bias are to be found in Paras. 145-156 and may be summarized as follows:

Re the bias arising from the known conflict with the Minster’s interests:

The evidence might show the Minister to be biased on the issue, but that did not prove, the Court said, that the adjudicator would be biased; “the presumption is that a decision-maker is impartial, absent evidence to the contrary”; “… [there is] no evidence the Minister’s comments influenced [the adjudicator]”; “[the adjudicator’s] evidence was that she was not influenced and that her position required that she ensure that not only she was not biased but also that she did not appear to be biased”. (Emphasis added.)

Re the ill-advised meeting:

“There is no evidence in the record that [the adjudicator] was actually influenced or that she deliberately acted unfairly in any way”; “there is a significant link in the chain of events which is missing…”; there is no evidence that the [adjudicator]… was influenced by or biased as a result of the meeting.  There is no evidence that the concerns raised [in the meeting] were conveyed … to [the adjudicator]”;

      Author’s Comments:

With great respect, none of this acknowledges the principles established in Valente and explained in greater particularity in Wewaykum (supra).

In stressing the lack of evidence that the adjudicator was in fact biased, or had acted in unfair way, and in relying on how she had been trained, and on what she understood her responsibilities to be, the Court ignores all of the instructions on bias  in Wewaykum. 

The  issue is only the apprehension of bias viewed objectively for three good reasons: (1) because it is often unrealistic or unwise to expect proof of actual bias; (2) because one cannot prove an unconscious bias, even though the existence of one disqualifies an adjudicator; and (3) because the apprehension of bias is the only relevant inquiry since the concern is the public’s confidence in the administration of justice.

This judgment on the bias issue does not in fact address what this adjudication looks like from a reasonable outsider’s perspective – how this decision might impact on the public’s confidence in the administration of justice in the immigration and refugee field – and does not consider at all the obvious likelihood of an unconscious bias in these circumstances.

In this judgment, the rule of law’s “fire” in the administrative justice system has been reduced to a mere illusion – to, as I say, an ignis fatuus: something that appears to be there but isn’t really.


Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top