Magical Thinking and Appearance-based Recusal

1. Philosophical Polarity – the “average” vs. the “ideal”

There have been two competing philosophical perspectives relevant to the legal idea(l) of reasonableness. One approach advocates a standard that is normative, one generally influenced by ethical values. The normative approach, which is predominant in the legal world, recognizes that the standard cannot be proven empirically or logically. This approach looks to reasonableness as reflecting a community's ideals and values, one that expresses the collective conscience of a community. 121

The competing view (labeled as positivist, empirical, or statistical) posits that the reasonable person is an ordinary “vanilla-type” creature, an all-purpose being reflecting the average citizen (“the average Joe”) and embodying an aggregation of beliefs and behaviors of the individuals in a community. 122 Such a composite approach is historically associated with its origins in statistics. As others have cautiously observed, the “average” approach, strictly applied, can implicate uncomfortable consequences. 123

Straddling the fence between these two camps is a legal philosophy that portrays the reasonable person as a hybrid in theory and practice. 124 In the prior discussion about the contrasting views of Justices O’Connor and Stevens in Capital Square regarding the identity of the reasonable observer, 125 there is a lurking issue whether the approach should be an idealized normative one, based on aspirational principles, or one that is more receptive to incorporating, at least in part, the empirical realities of a given community. 126 The normative-statistical dilemma will take on added significance with respect to reconceptualizing and customizing, to a degree, the reasonable observer heuristic in judicial ethics. 127

Such philosophical musings, sometimes abstruse, may be intellectually interesting. But they provide questionable practical guidance to the judicial decision-maker who must resolve disputes with clarity, practicality, and efficiency 128 Nevertheless, these competing philosophical perspectives are worthy of consideration because they may assist the decision-maker in identifying the appropriate values, points of view, sources of knowledge, and jurisprudential objectives in constructing and construing a context-and-fact dependent heuristic.

2. The Paradox of Objectivity

Objectivity in the law can be an overly romanticized aspirational concept. There are frequent references in caselaw that judicial reasoning is “objective.” Such a viewpoint is both idealistic and practical because it comforts the litigants and the public about the importance of judicial impartiality and the fair administration of justice, namely, that a jurist's personal preferences, values, or biases will/should not dictate the reasoning process. The reasonable person or reasonable observer becomes a valuable filtering mechanism for providing the appearance of objectivity and impartiality. At the same time, it provides an important reminder (to a jurist and the public) that personal values or views should not control the adjudicatory process.

But the concept of judicial objectivity requires a more nuanced assessment, as jurists and scholars acknowledge. Alan Calman has observed that what is missing from discussions of reasonableness is a basic understanding of human nature. 129 Prosser's analysis of the reasonable man concept admits that it implicates both the subjective and the objective. 130 Christopher Jaeger's analysis of the “empirical reasonable person” posits that the reasonable person's roots are empirical; but reasonableness is also intuitive and aspirational 131 In a legal zone that provides considerable unguided discretion, it is understandable that others have concluded that the use of reasonableness can disguise the lack of objective criteria and can operate as a disguise or a tool for judicial control that appears to defer to community standards. 132 The judicial task is especially challenging as Judge Kozinski noted in In re Bernard. 133 He has described the judge's philosophical dilemma as “this objective-subjective conundrum” wherein the jurist becomes both the interpreter and the object of interpretation. 134 As noted in In re United States, asking a judge to step outside himself and take the view of an objective outsider is a task that is “difficult even for a saint to do.” 135

Regardless of the context in which the reasonable person/observer standard is applied—torts, contracts, constitutional endorsement, or judicial disqualification—there remains an underlying concern about the ever-present danger of a judge's beliefs, values, predispositions, or bias imperceptibly compromising the apparent objectivity of decision-making, especially in circumstances when discretion is legally unguided. 136

3. Morality

The relationship between law and morality is a topic that has fascinated philosophical and legal scholars. H. L. A. Hart's classic exposition of the separation of law and morals explained that historically there has been a recognition that “the development of legal systems had been powerfully influenced by moral opinion, and, conversely, that moral standards had been profoundly influenced by law, so that the content of many legal rules mirrored moral rules or principles,” an historical causal connection that is not easy to trace. 137

The limited scope of this article precludes any extended philosophical discussion regarding the role of morality in the development of the law. 138 Suffice it to say that with respect to the reasonable person concept, others have observed a connection. In the realm of contracts, for example, Larry DiMatteo noted that the reasonable person's roots are in moral philosophy (Thomas Aquinas and Aristotle) and in a belief in virtues and right reason representing, in effect, a secularization of religious principles. 139 Given that the reasonable person is viewed as embodying the conscience of the community and is a personification of a community's ideal, it is natural that the reasonable person would assume a normative mantle. 140 The essential point is that the reasonable person/observer is plausibly imbued with normative, moral attributes. More importantly for our purposes, and regardless of the more general philosophical issues of law and morality, the reasonable observer in recusal matters should be recognized as a distinct construct that implicates moral/ethical considerations and aspirations. It is worth acknowledging that the essence of the reasonable observer metaphor in recusal is indeed virtue, in a secular sense, specifically, the civic morality of justice, judicial impartiality, and fairness.

4. Context

The issue of law's relation to morality raises the related and important factor of context. In his analysis of reasonableness and objectivity, professor Neil MacCormick stressed that the task of interpreting “reasonableness” is contextual, involving the identification of values, interests and the like that are relevant to the particular focus of attention, which depends on the type of situation, the relationship at issue, and the governing principles and rationales for the branch of law at issue. Reasonableness is necessarily a context-driven concept. 141 Justice O’Connor in Capital Square explained that the application of her reasonable observer-endorsement test depended on a sensitivity to the unique circumstances and context of the particular challenge. 142 Other commentators caution that one must be careful in applying the reasonable person concept beyond traditional legal realms. 143 One thus needs to acknowledge the special context of appearance-based recusal, particularly with respect to the underlying values and concerns that would be relevant to the judicial interpretation of the reasonable observer. Context—the public's perception of judicial impartiality in the administration of justice vis à vis the particular facts and circumstances of a case—is all-important. The fact, for example, that the Supreme Court has recently abandoned the reasonable observer approach in Establishment Clause cases (in favor of a history-and-tradition test) 144 does not dictate a similar result in appearance-based recusal jurisprudence given the fact we are faced with the unavoidable task of carefully explaining and applying a paramount ethical standard that textually incorporates the metaphorical reasonable observer.

1. Procedural Preliminaries: Allocation of Benefit and Burden

Inasmuch as impartiality is a foundational value in our justice system, a disqualification challenge represents a weighty and an emotionally precarious challenge to the judicial system and the judge. Given the gravity of the matter, strict guardrails have been established to prevent frivolous claims or tactical manipulation of the judicial process. These procedures impose a burden (on the petitioner) and a benefit (on the jurist).

A petitioner who claims actual or apparent bias must present a claim with factual specificity. Vague, conclusory, unverified, or unsupported allegations or feelings are insufficient to satisfy the petitioner's evidentiary hurdle. Thus, general allegations of animus, as well as speculation or innuendo, cannot satisfy the evidentiary burden. Courts will reject recusal challenges when they are based on “mere” conjecture or suspicion. 191 As one court has noted: “…disqualification should not be allowed on the bases of rumors, innuendos, unsupported allegations, or claims that like blind moths, flutter aimlessly to oblivion when placed under the harsh light of full facts.” 192

The petitioner faces another burden. A challenge to a jurist's actual or apparent impartiality must meet the obstacle of a presumption that strongly benefits the challenged jurist. 193 The presumption is long-standing, recognized in the eighteenth century as vital to the common law system, which adopted a restrictive approach to disqualification. 194 Disqualification caselaw in the United States routinely asserts that a jurist is presumed to be competent and to possess integrity. 195 The burden to disqualify a judge and overcome the presumption is viewed as a heavy one. 196 Looking at the presumption from an angle other than competency and integrity, one court started its disqualification analysis with a “presumption against disqualification,” which arguably reflects the presumption's true impact. 197 Similarly, another jurist has observed that great deference must be given to a trial judge facing a recusal challenge, a sentiment that permeates disqualification jurisprudence. 198 The presumption is a significant hurdle for the litigant.

Aside from the issue of providing a challenged jurist with a procedural advantage in a recusal challenge, the presumption generates other concerns. Judge Easterbrook noted:

Yet, drawing all inferences favorable to the honesty and care of the judge whose conduct has been questioned could collapse the appearance of impropriety standard under sec. 455(a) into a demand for proof of actual impropriety. So although the court tries to make an external reference to the reasonable person, it is essential to hold in mind that these outside observers are less inclined to credit judges’ impartiality and mental discipline than the judiciary itself will be. 199

This jurisprudential concern leads to another important issue. What is the actual effect of the presumption? Does it tilt the scales of justice in the jurist's favor? For example, is it applied at the initial stages of litigation or throughout the litigation, thus providing a tactical advantage for the judge and a procedural burden on the petitioner? There is no clarity in recusal caselaw. One suspects that the presumption operates to benefit the jurist throughout the disqualification litigation. Presumptions can be conclusive or rebuttable. Presumptions are created for reasons of convenience, fairness, or policy. 200 One view is that, if evidence is produced to rebut the presumption, the presumption is utterly destroyed and disappears (the so-called “bursting bubble” theory) 201 even if the decisionmaker disbelieves the countervailing evidence. 202 The weight of authority is that the presumption, however, does not have any effect on the persuasion burden; it merely shifts the production burden; litigants challenging a jurist's qualification must still prove their case. 203 In the reasonable observer context, commentators have been critical of the application of the presumption, suggesting that the presumption be re-considered and re-calibrated. 204

2. The Reasonable Observer --The Enigmatic Wisdom Whisperer

To understand what and how the reasonable observer perceives, it is necessary to ascertain who the reasonable observer represents. A transcribed administrative conference discussion between two justices of the Michigan Supreme Court, regarding Michigan's then recently amended rules of disqualification, 205 highlights a conceptual consternation:

Justice Hathaway::  If there is an appearance of impropriety, then you cannot sit on the case.

Justice Young::  And from what perspective is the appearance of impropriety? Is it a subjective standard? Is it an objective standard?

Justice Hathaway::  I haven’t thought through all of that to be honest with you, to answer you here. 206

The justices’ perplexity is understandable because, in assessing the appearance of impropriety, a jurist is placed in an awkward, perhaps cognitively untenable, position. As one jurist observed: “An objective standard creates problems in implementation. Judges must imagine how a reasonable, well-informed observer of the judicial system would react. Yet the judge does not stand outside the system.” 207 In quoting from another case, Judge Kozinski remarked: “Because the judge must apply the disqualification standard [of section 455(a)] both as its interpreter and object, the general standard is even more difficult to define. [There is a] philosophical dilemma created by this objective-subjective conundrum.” 208

From the theoretical perspective at the legal baseline, however, courts have recognized that the reasonable observer should not be the judge—the reasonable observer must be a lay person. 209 One court expanded the traditional perspective by stating that “the question of reasonableness ought to be approached from the viewpoint of the party to the action, not of that famous fictitious character, the reasonable man.” 210 Since the observer's perspective is theoretically an objective one, it should not embody the personal values, philosophy, or viewpoint of the jurist tasked with applying the standard, especially if the jurist is the object of the ethical inquiry. This approach is consistent with Anglo jurisprudence. 211 The difficulty, however, is that the reasonable observer remains an abstraction and inevitably leads to a deeper dilemma, i.e., what are the attributes of the imaginary reasonable observer? Analytical clarity is problematic. 212

In the negligence field where the reasonable person came to maturity, Dean Prosser remarked that the level of knowledge, including minimal requirements, ascribed to the reasonable person is one of the most difficult issues to assess. 213 In disqualification cases, the commonplace expressions are that the reasonable observer is one who is “informed” of all the surrounding facts and circumstances; a thoughtful person, but not hypersensitive or unduly suspicious; one who is knowledgeable and objective. 214 The reasonable observer is viewed as “the average person on the street.” 215

Difficulties arise when the “knowledge” and “fully informed” aspects of the reasonable observer are examined more closely. The definitional quandary brings to mind the differing views between Justices O’Connor and Stevens in Capital Square about how much knowledge (of the history and context of the community) should be imputed to the reasonable observer. 216 Notwithstanding the moniker of the reasonable observer as an “average Joe,” the observer has been identified in disqualification cases as someone who is outside the judicial system or even unfamiliar with it, one less inclined than the judiciary itself to credit a judge's impartiality. 217 These characterizations may reflect an attempt to emphasize a more visible, confidence-inspiring, wall of separation between the observer and the jurist/judicial system. Clearly, black-letter law repeatedly states that the reasonable observer is informed, not uninformed, knowledgeable of and understands the facts and circumstances of the matter. 218 But what do these attributes mean? Some cases have imposed a responsibility on the hypothesized observer to examine the facts, the record, even the law and judicial practices, 219 going so far as to impose a quasi-legalistic perspective onto the reasonable observer. 220

Other issues about the knowledge and point of view of the “fully informed and objective” observer arise. Often, such facts may be hidden from public view and are not readily ascertainable—for example, the association of a judge's law clerk with one of the parties or counsel, the potential economic interest or civic activities of a judge's spouse, an ex parte conversation, a financial gift or contribution, or a troubling past social media post. Such “private” facts may indeed be relevant to an appearance-based challenge. Although caselaw states that the recusal inquiry is tied to knowledge of facts in the public domain, 221 appearance-based recusal may require the examination of not readily ascertainable facts. These private facts eventually become public when they are made part of the official record. 222 A legitimate concern arises, however, when such private facts represent insider information and are used to boot-strap a refusal-to-recuse decision. 223

Lastly, against this tableau of analytically diverse perceptions of the reasonable observer, one returns to the fundamental issue of what the reasonable observer heuristic is (or is not) capturing. Philosophically, there has always been a tension in how the reasonable person heuristic is applied. As noted previously, 224 should it simply embody the “average” of a society? Or is there a normative or idealized component to the construct? The answer may be both. 225

Justice O’Connor's vision of the reasonable observer (in religious endorsement cases) had always been a challenging one. In Capital Square, Justice O’Connor disavowed any focus on “actualities,” preferring to base her heuristic on a “collective standard,” similar, she said, to the reasonable person in the law of torts. 226 Justice O’Connor acknowledged that the fictional metaphor in tort represents a “community ideal of reasonable behavior.” 227 Prosser also described the reasonable person as the “personification of a community ideal.” 228 The personification, however, goes further. From Justice O’Connor's perspective, the reasonable observer was viewed as aware of the history and context of the community and the forum in which the religious display appears. 229 In disavowing consideration of “any person” or “some people,” 230 Justice O’Connor applied a metaphor that relies on both an average and an idealized-normative personification of the community. The construct is fundamentally theoretical and abstract, intuitively (i.e., subjectively) based, with no apparent connection to an empirical thread. 231 The reasonable observer is, in effect, an abstract portrait painted with a broad brush. 232

The application of Justice O’Connor's formulation of the metaphorical reasonable person/observer in Establishment jurisprudence revealed some underlying infirmities of the heuristic. Notwithstanding the demise of Justice O’Connor's heuristic, the critical questions asked by commentators remain relevant for our purposes: Whose perception controls? 233 If the reasonable person represents an “average,” what is it an average of? 234 With respect to such concerns, the application of the heuristic in religious endorsement cases had been criticized as being both under-inclusive and over-inclusive. 235 Echoing Justice Stevens’ assessment in Capital Square, 236 such commentators opined that the reasonable observer heuristic often favored a majoritarian point of view; it could be insensitive to non-majoritarian or minority perspectives (cultural and individual). Another commentator, Paula Abrams, had said that the reasonable observer is a formalist characterization, devoid of real human reactions, an empty suit that lacks humanity, a standard that undermines the value of inclusion. 237 It is in this respect that the reasonable observer heuristic presents a significant qualitative difference regarding its application in Establishment and disqualification cases. Establishment-endorsement cases inevitably involve the application of the heuristic in relation to particular constitutional values. 238 Constitutional terrain is simply different – due process and religious liberty, for example, involve concerns and values distinct from the subject matter of litigation in which recusal is raised. 239 Thus, it is important to consider the context of the recusal challenge when applying the heuristic's requirement of reasonableness. 240 In disqualification cases, for example, the factual context of the recusal challenge is unrestricted and can be wide-ranging. Impartiality challenges can be linked to many factual variables: religion, gender, race, sex, ethnicity, and political issues. Despite the contextual differences, as in endorsement cases, there is always the danger of an anti-majoritarian bias or insensitivity seeping into disqualification assessments. In a multi-cultural society, the reasonable observer in disqualification cases should not be considered in majoritarian or statistical terms, even if such an endeavor were possible. Disqualification cases are qualitatively distinct in context because the precept of judicial impartiality, and the appearance thereof, are values that are foundational to the rule of law and the decisionmaker's (and judicial system's) integrity and credibility. In short, the interpretation and application of “the reasonable observer” heuristic, integral to the text of the ethical disqualification mandate, require a cautious, customized, and contextually sensitive approach.

Metaphors (like the reasonable person or observer) are meant to assist us in thinking and reasoning. 241 They expand our perceptual horizons. In law, metaphorical devices should serve to promote rationality, analytical predictability, and the appearance of adjudicatory fairness. In the analysis and application of the reasonable observer metaphor in recusal caselaw, however, the lack of the heuristic's clarity exposes a troubling uncertainty about the “wisdom whisperer”.

3. Semantics and the Spectrum of Belief

As Mephisto advised in Faust: “Put your trust in words/ They’ll guide you safely past doubt and dubiety.” 242 Similarly, one finds another literary character, Alice in Wonderland, created years after Faust, raising a fundamental linguistic dilemma with Humpty Dumpty. In response to Humpty Dumpty's assertion that “When I use a word it means just what I choose it to mean – neither more nor less,” a puzzled Alice says: “The question is whether you can make words mean so many different things.” 243 An examination of disqualification jurisprudence in the United States reveals the wisdom of that observation. 244

The over-arching disqualification 245 standard in the United States is that a jurist must disqualify when the jurist's “impartiality might reasonably be questioned.” It is a specific standard, reified in federal and state statutes and judicial codes, similar in principle to, but distinct in form from, its counterpart in common law countries (which rely on general principles of apparent impartiality and the appearance of bias). In the U.S. standard, the verb “might” acts as the fulcrum of implementation. The operative word is arguably one of lexical simplicity. In common parlance, the modal verb “might” occupies a position within a spectrum of predictability and certainty; it is an expression that connotes possibility. 246 For example, if the weather forecaster states that “it might rain,” rather than “it would rain” tomorrow, one would interpret the former forecast as more hospitable to the planning of an outdoor event. 247 While philosophical or lexical interpretations may engender complexity, confusion or ambiguity, 248 the common understanding of the two modal verbs (might and would) reflects a substantial epistemological difference --- from possibility to probability. 249 Unlike weather forecasting that relies on objective atmospheric criteria and mathematical calculations, however, recusal assessments present greater difficulty and risk of error because they depend on the subjective-objective 250 analysis of the dauntingly imprecise ingredient of “reasonableness.”

The crux of this Article's section is that U.S. recusal jurisprudence presents a perplexing example of the lack of analytical clarity regarding the meaning of the appearance recusal heuristic and the applicable evidential threshold for disqualification. Specifically, there is a disturbing divergence in disqualification jurisprudence between the specific terminology of the ethical mandate (disqualification is required when a judge's “impartiality might reasonably be questioned”) and its application in concrete cases -- a divergence that ultimately undermines the fundamental value that justice must satisfy the appearance of justice. Remarkably, except for the occasional perceptive observation by others of the conceptual ambiguities, 251 there has been a lack of analytical attention regarding the critical issue of the evidential threshold of belief in appearance-based disqualification. What is the judicial lens? As Richard Flamm pointedly asks: Does the disqualification standard embody a notion of conceivability or certainty? Flamm identifies the linguistic and conceptual conundrum in terms of “definitely would question” or “might conceivably do so.” 252 A general exposition of the caselaw leads Flamm to conclude that there is a split of opinion. As he notes, courts have rarely squarely trained their attention on this issue. 253 The ad hoc and non-analytical approach to judicial disqualification, in the absence of any authoritative guiding principles, has contributed to a perception of inconsistency and ambiguity. Nevertheless, it is beneficial to determine if there are discernable patterns emanating from the collective judicial conscience. 254

From a wide-angle perspective, U.S. caselaw seems to slip and slide from the lower modal standard (“might”) to a higher conclusory “would” – the latter, in Faustian parlance, safely guiding the decision-maker from doubt and dubiety. 255 To say that a reasonable observer “might” reasonably question a jurist's impartiality is significantly different from concluding that a reasonable observer “would” (but, more often in reported cases “would not”) question the jurist's impartiality – again, predicated on a subjective (or magical) assessment of the hypothetical perception of the hypothetical reasonable observer.

U.S. caselaw reveals an analytical approach that is less solicitous to appearance-based recusal, one that is in tension with the ordinary and clear text of the standard. Often the modals “might” and “would” are blithely used interchangeably in opinions (and even in a single opinion). For example, in rejecting countervailing considerations of administrative inconvenience and expense of a re-trial in a convoluted multi-party diversity action (that required 33 days of trial), one court adopted a hard line approach toward the trial judge's failure to disqualify, stressing the importance of protecting the judiciary from any hint of the appearance of bias. 256 Nevertheless, the court's use of words is noteworthy, when it said: “The judge should consider how his participation in a given case looks to the average person on the street. Use of the word ‘might’ in the statute was intended to indicate that disqualification should follow if the reasonable man, were s/he to know all the circumstances, would harbor doubts about the judge's impartiality... [then noting that] [A] reasonable person might very well question the judge's impartiality.” 257 It's a head-spinning analysis. Although such interchangeable use of “might” and “would” within opinions is common, 258 occasionally one does see in other cases an analysis and result that are faithful to the precept's modal “might.” 259

In addition, courts will frequently couple the outcome-determinative modal verb with qualifiers that make the advocate's burden more onerous. In the application of the relatively simple five-word recusal standard (i.e., the judge's “impartiality might reasonably be questioned”), courts exercise considerable interpretative latitude and creativity in the assessment of the risk of perceived partiality. Courts have imposed various conditions onto the “might” appearance standard, including: “significant doubt;” 260 “serious doubt;” 261 “significant risk;” 262 “substantial doubt;” 263 or “substantially out of the ordinary.” 264 Some cases will also identify a more burdensome evidential standard. In United States v. Nixon, the court, beginning with the protective presumption of impartiality, noted that the moving party must demonstrate by “clear and convincing evidence” that a judge has conducted himself in a manner supporting disqualification. 265 Although courts may sometimes frame the standard of review in terms of reasonable doubt, 266 one court had to specifically disavow a “beyond a reasonable doubt standard,” calling such strict language in prior caselaw a “minor oversight.” 267

Lastly, there are instances when a disqualification challenge has been rejected despite an acknowledgment that there may indeed be merit to a reasonable person's questioning the jurist's impartiality. 268 In Parker v. Connors Steel, 269 a complicated labor dispute case involving allegations about the conflicting participation of the judge's law clerk in the decisional process, the court seems to have turned the appearance-based recusal standard on its head when it rejected a disqualification challenge and found harmless error, saying:

To the extent that public confidence has already been undermined, we do not believe that granting relief in this case will change the public's perception in any appreciable way. Such harm cannot be remedied by vacating the district court's decision and reassigning this case to a different judge. In fact, if we reverse and vacate a decision that we have already determined to be proper, the public will lose faith in our system of justice because the case will be overturned without regard to the merits of the employees’ claims. Judicial decisions based on such technical arguments not relevant to the merits contribute to the public's distrust in our system of justice. 270

It is impossible to identify the impetus (psychological or jurisprudential) for the imposition of a higher standard in these appearance-recusal cases. Perhaps an aversion to the challenging and vague appearance-based standard; or an unconscious preference for (or comfort in) an actual prejudice standard; 271 or, from a speculative sociological perspective, 272 the unexamined semantical habits or shared understandings in the judicial community's zeitgeist—these may explain the more restrictive (i.e., the higher evidential “would”) approach in appearance-based recusal cases.

In any event, such varying adjectival adhesions, increasing the procedural and evidential burdens imposed on a petitioner, effectively transmogrify the appearance-based recusal standard, create analytical confusion, and increase the risk of erroneous and unfair decision-making. The ultimate risk is that the public's perception of justice and its trust and confidence in the judicial system are jeopardized.

1. The Reasonable Observer: Identity

Commentators, including Supreme Court justices, 349 have advocated for a more realistic, sensitive, and nuanced conception of the reasonable observer. 350 As others have suggested, the reasonable person/observer is a heuristic that should reflect social (public) meaning; the heuristic should acknowledge and incorporate the real possibility of multiple personae. 351 Relevant to a broader, more flexible heuristic is the recognition of the futility and undesirability of trying to achieve an idealized, unattainable consensus. 352 The flexibility of this approach makes philosophical and jurisprudential sense if one considers the fundamental nature of the reasonable observer. In response to the persistent philosophical debate whether the reasonable person, as the designated representative of a global community (“the average Joe,” so to speak) is more statistical (i.e., average) or normative (i.e., the embodiment of an ideal or community values), commentators have favored the latter. A purely statistical approach is viewed as empirically impossible inasmuch as we lack objective means to reduce human beings or their beliefs to a single number, metric, or trait. 353 The statistical approach, in its attempt to generalize reality, presents the danger of being over- or under-inclusive; 354 in a sense, conceiving of reasonableness as an average or composite of multiple characteristics results in an unrealistic leveling of reality – it captures too much or too little, and thus can be viewed as exclusionary, a particularly troublesome analytic when placed in the context of ethics and justice. 355 Additionally, supportive of a more flexible and recusal-sensitive approach to the reasonable observer heuristic is the fact that the heuristic is applied to the ethical domain of judicial impartiality, a secular value that ultimately reflects the ethic of caring for the interests of others, 356 a viewpoint that is compatible with the classical notion of the reasonable person. 357

The recognition of the interests of others, when relevant, should guide the formulation of the reasonable observer heuristic. The high court of Australia addressed the importance of considering the impressions of the public and parties in applying the reasonable observer (a/k/a fictitious bystander) heuristic:

It is their confidence that must be won and maintained. The public includes groups of people who are sensitive to the possibility of judicial bias. It must be remembered that in contemporary Australia, the fictitious bystander is not necessarily of European ethnicity or other majority traits.” 358

Similarly, in identifying and applying the objective test for apparent bias, the High Court of South Africa acknowledged: “In a multicultural, multilingual and multiracial country such as South Africa, it cannot reasonably be expected that judicial officers should share all the views and even the prejudices of those persons who appear before them.” 359 In a racially-charged case, involving a white police officer's arrest of a Black 15-year old who had allegedly interfered with the arrest of another youth, the Supreme Court of Canada applied its reasonable apprehension of bias test with the following caution: “Judges must be particularly sensitive to the need not only to be fair but also appear to all reasonable observers to be fair to all Canadians of every race, religion, nationality and ethnic origin.” 360

The preceding commentary is relevant to the symbolic and practical issue of whose voice does the judge channel when conjuring the metaphorical reasonable observer. Joanna Grace Tinus, advocating a fine-tuning of the heuristic, has remarked that “…the objective nature of the [reasonable person] standard has been undermined by relying on a standard of reasonableness that tends to reflect social norms and particular prevailing ideas of particular classes of individuals.” 361 Others have focused their criticism on the fact that the reasonable observer heuristic suffers from an inherent majoritarian point of view, sometimes characterized as the “individuation problem.” 362 Associating the reasonable person with a majoritarian point of view, for example, had been recognized as a serious defect of the heuristic (as previously applied in America's religious endorsement cases) given the potential impact on minority populations. 363 Jesse Choper, for example, had recommended that religious minority interests should be part of the calibration. 364

The recognition and incorporation of multiple perceptions, when appropriate and feasible, would promote greater jurisprudential sensitivity and clarity. Decision-making could be enhanced, for example, by taking a debiasing “external assessment approach.” As Richard Re notes:

Thus, a court could attempt to assess and take account of the views of other actors, even when the court itself is “internally” certain that the other actor's reasonable view is incorrect. Scholars have labeled this basic approach an external assessment of ambiguity, by which one interpreter attempts to predict or imagine how other interpreters would resolve a particular issue. 365

This external assessment of ambiguity approach, in which the identity of the perspective plays a key role, is believed to enhance analytical clarity and predictability. Such a mode of interpretation may be more appropriate when there is limited information and the governing perspective is that of an actor other than the deciding court, 366 conditions that apply in the recusal context. Ward Farnsworth explains that the external assessment approach focuses on how ordinary readers would view an ambiguous issue. 367 Noting that internal assessments about ambiguity are dangerous because they are easily biased by strong (sometimes unconscious) policy preferences, Farnsworth observes that the “external estimates of ambiguity, while sometimes inaccurate, are nevertheless more accurate than internal judgments when measured by the amount of agreement readers are able to reach about a statute [or text].” 368 In reference to the task of interpreting an ambiguous statute, he states:

The external perspective…can serve as a useful heuristic in such cases where the clarity of a text is open to question, especially in areas of law where parties – or “ordinary readers” of the legal text in question – have a strong interest in notice. The external standard is a valuable corrective to the serious risks of bias that attend the more usual task of simply asking whether a statute seems clear to oneself. 369

The external approach in interpretation is a sensible one given that, as Christopher Brett Jaeger has noted, there is a distinction between legal reasonableness and lay reasonableness. Academic or theoretical discussions of the reasonable person, whose roots are empirical, are often divorced from the reality of how lay decision-makers encounter, understand, and apply the standard; it is an issue, he says, that deserves more attention. Jaeger posits that law should, as a normative matter, track the lay conception of justice and should mirror popular intuition.” 370

This analytical backdrop leads to the fundamental practical question as to the identity of the voice(s) of the reasonable observer. As emphasized by the high courts of the United Kingdom, Australia, and South Africa, 371 the fair-minded reasonable person is a lay person, not the judge—notwithstanding the reality that some subjectivity will inevitably seep in because a human (the judge) is the medium for interpretation. Likewise, the perceptions of the public and parties, while not determinative or controlling, are worthy of consideration in the formulation of the heuristic since the confidence of the litigants and the parties in the judicial system is fundamental. 372 Finally, while more difficult to assess, the reasonable sensitivities and perceptions of apparent bias, shared by identifiable segments of the population, should be considered if their “voices” have relevance to the issues in the proceeding given the over-arching policy objective of impartial decision-making. 373

The identity of the reasonable observer is difficult, yet fundamental, to the integrity of the decision-making process. The issue of the hypothetical reasonable observer raises philosophical, jurisprudential, and pragmatic concerns. Since the reasonable person/observer question must be rooted in the realities (albeit speculative) of the lay observer, an empirical assessment would be a rational way to proceed. But how? While recognizing that the reasonable observer question is ultimately one of law, 374 and is not determined by a simple calculation of votes, Jessie Hill concludes that an empirical consensus is difficult (albeit inappropriate) to attain and, ironically, runs the risk of supporting a discriminatory majoritarian point of view. 375 Nevertheless, she posits that the reasonable observer's task (i.e., the determination of public or social meaning) can be approached by evaluating all relevant information, 376 similar to the suggestion made by the Canadian Supreme Court, which stated:

Judicial inquiry into context provides the requisite background for the interpretation and the application of the law. An understanding of the context or background essential to judging may be gained from testimony by expert witnesses, from academic studies properly placed before the court, and from the judge's personal understanding and experience of the society in which the judge lives and works. This process of enlargement is a precondition of impartiality. A reasonable person would see it as an important aid to judicial impartiality. 377

If information is available, and if the task is reasonably feasible and evidentially relevant, the process of enlargement should be considered. Doing so would make the reasonable observer heuristic in appearance-based judicial ethics more principled, jurisprudentially sound, and responsive to the changing realities of contemporary society's pluralism.

2. The Reasonable Observer: Imputation of Knowledge

Anglo-American jurisprudence identifies the metaphorical reasonable observer in generalities: fair-minded, reasonable, thoughtful, aware of the relevant facts and circumstances, and informed. As the prior discussion has indicated, the “informed” attribute has generated a considerable variety of opinion about the reasonable observer's level of knowledge and information. 378 How “informed,” “well-informed,” “fully informed,” or “knowledgeable” must the reasonable observer be? Discussion among judges and academics about the cognitive capacity and imputation of knowledge has occurred in two different legal contexts: constitutional religious endorsement and disqualification. Justice Stevens was particularly troubled by Justice O’Connor's more sophisticated formulation of the reasonable person heuristic as previously applied in the religious endorsement context. For Stevens, the legal construct of the reasonable observer unrealistically represented a well-schooled jurist and a personification of a community ideal who possessed a high level of legal and historical knowledge. 379

Despite the different contexts (i.e., constitutional religious endorsement and rule-based judicial ethics), the basic jurisprudential challenges about the “informed” reasonable observer are similar. American commentators concluded that the heuristic (as it had been applied in the constitutional religious endorsement context) presented a highly problematic, over-idealized, unrealistic caricature regarding the imputed level of knowledge. 380 Common law commentators have also expressed their concerns about the “informed” attribute regarding their recusal jurisprudence. As noted, some common law countries have imposed a more elaborate or rigorous standard of the informed attribute. 381 That approach has been criticized. 382 Although expressing his displeasure with the “artificial” and “unworkable” reasonable lay observer heuristic, and favoring a return to a judge-centric approach, Professor Olowofoyeku, noted a trend that common law courts were imbuing the informed observer with increased knowledge and understanding so courts can reach a “right outcome,” which he says is inconsistent with the rationales for interposing a hypothetical lay person to judge the appearance of bias. As such, he notes, “this impartial observer might as well be a judge.” 383 Similarly, critical of imbuing the reasonable person with insider information and the workings of the judicial system, two commentators have viewed the application of a higher standard as a way for courts to justify their refusal to recuse. 384 In their view, this interpretation of the informed observer augments the significance of the judge's sensibilities, hence subjectivity, and plays an important role in compromising judicial integrity and the apparent bias test. 385 Simon Atrill, a proponent of a more nuanced observer test that emphasizes a balance of policy interests, likewise, viewed the imputation of a higher-level of knowledge as effectively facilitating a return to the Gough standard in which reasonableness is seen and judged through the eyes of the jurist. 386 As the High Court in Singapore observed: “It is also why it would be a mistake for a court to simply impute all that was eventually known to the court to an imaginary reasonable person because to do so would be only to hold up a mirror to itself.’ 387

The Australian judicial system has stressed the importance of adopting realistic criteria for the variously described fictitious bystander. As the High Court of Australia explained: “Obviously, all that is involved in these formulae is a reminder to the adjudicator that, in deciding whether there is an apprehension of bias, it is necessary to consider the impression which the same facts might reasonably have upon the parties and the public.” 388 To that end, Australia often omits the “informed” attribute in applying the reasonable bystander heuristic. 389 For example, as noted in Johnson, the bystander is described as fair-minded and reasonable, neither wholly uninformed or uninstructed about the law in general or issues to be decided, knowledgeable about commonplace things, and possessing basic common sense regarding the process of adjudication and the judicial-legal profession, one who is neither unduly sensitive or suspicious. 390

The High Court of England and Wales noted that the fair-minded observer cannot be ascribed all the knowledge and, indeed, assumptions of a trained judge, adding “The fair-minded and informed observer can be assumed to have access to all the facts that are capable of being known by members of the public generally, bearing in mind that it is the appearance that these facts give rise to that matters, not what is in the mind of the particular judge or tribunal member who is under scrutiny.” 391

How one describes—or embellishes—the attributes of the reasonable observer can be, knowingly or unwittingly, outcome-determinative. For conceptual and interpretive clarity, the reasonable observer should not be imbued with unrealistic or unnecessary qualities that threaten to convert the reasonable observer heuristic into a subjectivized judge-centric standard that muddies the focus of the standard (the objective and fair-minded lay member of the community) or undermines the standard's fundamental values (appearance of impartiality, public trust and confidence). If an “informed” attribute is deemed necessary, then it should be a simple one, connected to the relevant facts and circumstances of the case – an attribute that supports the desired qualities of being thoughtful, fair-minded, and reasonable. Simply put, how “informed” must one be to make a commonsense, reasonable assessment of a jurist's apparent impartiality? 392

1. The Judicial Transmogrification of a Clear Mandate

Identifying the voice and attributes of the reasonable observer is, as the high courts of Singapore and Australia recognized, the portal to understanding and applying a critical element of the apparent bias heuristic, viz., the level of scrutiny applicable to the assessment of the reasonableness of a lay observer's perception of bias. 393 As the preceding sections demonstrated, 394 the common law countries have engaged in semantic struggles to identify the appropriate level of proof for assessing apparent bias: from “real danger,” to possibility, to likelihood, to probability -- all considered in relation to the metaphorical observer's enigmatic manifestations (such as “apprehension” or “suspicion”). The labyrinth of language employed in the search for understanding and consensus has been Faustian. 395

Commentators and jurists in the United States, on the other hand, have avoided (intentionally or unreflectively) such semantical quicksand. The approach has been devoid of meaningful analysis in the interpretation and application of the reasonable observer heuristic's “might reasonably be questioned.” The modal verb “might” is the outcome-determinative fulcrum of the standard. 396 “Might” and “would” are distinct terms. 397 Yet, because of the lack of interpretive guidance, there has been confusion regarding the level of probability required: does it connote, as one commentator has observed, a higher level of certainty (“would”) or lower a lower level of conceivability (“might”)? 398 In terms of American recusal principles and practice, the question raises an important jurisprudential issue. Within a fluid spectrum of uncertainty, what is/should be the appropriate level of belief and evidential proof ? 399 The dilemma of how to allocate the burden is exacerbated when information and human cognitive abilities are limited. 400

Notwithstanding such constraints, the law has attempted to calibrate certitude, although, as one commentator has noted, remarkably no one has ever formulated an adequate model for applying the standards of proof. 401 Kevin Clermont notes: “The epistemological aim of evidence law is that the factfinder should construct a belief that corresponds to the outside world's truth. Probability thus reflects a measure of the chance of that correspondence existing between finding and reality.” 402 The traditional method of legal reasoning is through imprecise probabilities. Civil law, for example, assigns evidential burdens through various perspectives such as preponderance of the evidence or clear and convincing evidence. 403 Criminal law has adopted additional calibrations, such as reasonable suspicion, probable cause, and beyond a reasonable doubt. 404 The Singapore High Court, for example, placed the “imaginary scales of justice” in distinctly impressionistic terms: doubt, suspicion, likelihood, and more-likely-than-not. 405

Academics have not been able to resist the allure of positing alternative theories and methods to identify degrees of probability and certitude. 406 Evidential calibrations are inherently imprecise and unquestionably implicate a high degree of intuition and subjectivity in the decisionmaker. Attempts have been made to identify a hierarchy of standards of proof within the realm of traditional probability. Kevin Clermont, for example, disfavors quantification and has offered the following scale (“categories of uncertainty”) regarding decision-making: 407

1. Slightest Possibility

2. Reasonable Possibility

3. Substantial Possibility

4. EQUIPOISE

5. Probability

6. High Probability

7. Almost Certainty.

Clermont notes that a higher standard is a way to inform the factfinder that the burdened party must provide a stronger showing of probability; a better way to envisage the whole scale of likelihood, he says, is as a set of fuzzy categories, or coarse gradations, of likelihood. 408

Nevertheless, there is a gravitational pull to seek greater clarity and certainty through the assignment of more specific metrics, although judges reportedly eschew numerical or percentile interpretations. 409 Ronald Bacigal, for example, has reformulated the levels of certainty into five categories by assigning the following statistical benchmarks: 410

1. Slight Possibility (1% to 10%)

2. Reasonable Suspicion (20% to 40%)

3. Fair Probability (40% to 49%)

4. More Likely Than Not (51%)

5. High Probability (80% to 100%)

Irrespective of the challenge of identifying and assigning probabilistic numbers to the standards of proof, Clermont, for example, acknowledges that the law allows recovery upon much less than a 50% showing of probability. 411

The discussion of heuristic calibration takes one closer to an understanding of what should be a potentially more principled and rational understanding of the disqualification standard's “might.” To do so, there is a need to expand the horizons by considering two related, but distinct, standards of proof that are applied in the criminal law context: probable cause and reasonable suspicion. 412

In the context of Fourth Amendment law, 413 “probable cause” is not what it appears to be. Probable cause is not synonymous with “probably.” Probable cause signifies more than bare suspicion; nor does it require resolution of evidence according to a preponderance of the evidence or the more-likely-than-not standard. 414 Probable cause is understood as requiring a reasonable ground for belief. 415 Recognizing that probable cause is a fluid concept not easily reducible to a neat set of legal rules, Kiel Brennan-Marquez notes that the Supreme Court's reasoning in probable cause tracks the plausibility model of suspicion. 416

This discussion takes us to the U.S. concept of reasonable suspicion, which has its roots in Terry v Ohio. 417 Craig Lerner noted that, in quantitative terms, and in comparison to probable cause on the spectrum of probability, reasonable suspicion amounts to far less than 50%. 418 With its origins in English law and as explained in the Canadian case of R v Kang Brown, “A ‘reasonable’ suspicion means more than a mere suspicion, and something less than a belief based on reasonable and probable grounds.” 419

For our purposes, aside from its relatively lower-level quantitative aspect, the notion of suspicion is a fluid concept that reflects practical considerations of everyday life. 420 Regarding both standards (reasonable suspicion and probable cause), the requirement of a narrative, factual explanation based on the totality of circumstances presented is important. The standards are concepts designed to explain, not predict. Brennan-Marquez notes that the Supreme Court has long understood probable cause and reasonable suspicion in explanatory terms, i.e., requiring articulation of data and information supporting one's inference or conclusion. 421 Essentially, identifying the governing standard of scrutiny with clarity, in conjunction with the requirement of factual articulation, would help to constrain discretion and subjectivity.

2. Reasons That Support a Clear and Strong Disqualification Standard

There is a need to re-interpret the appearance-based disqualification standard in a manner that re-balances the equation away from popular notions of probability or certainty. In doing so, we need to acknowledge the current unreflective jurisprudential approach and the importance of principled, analytical clarity. The operative disqualification standard—when a judge's impartiality might reasonably be questioned—should be interpreted more carefully and less restrictively than it has been. The critical issue is how one interprets and applies the modal “might,” as modified by “reasonably.” In consideration of the preceding discussion about levels of belief, the appropriate level of scrutiny should be reasonable suspicion—not likelihood or probability. The following reasons justify such an approach.

a. Textual and semantic fidelity, ordinary usage: The glaring aspect of the appearance standard is that “might” is not synonymous with, and does not have the same semantical meaning of, “would.” The modal “would” is utilized frequently, without explanation or elaboration, in U.S. disqualification caselaw. To be clear, there is no semantic ambiguity in the disqualification standard's specification of “might.” Whether “would” was considered by the drafters as an option, we do not know; in any event, the drafters specified “might.” Although it is impossible to discern the actual intent or state of mind regarding how one uses or interprets language, we can presume that words are used in a way that is consistent with their plain meaning—similar to how we approach and differentiate actual from apparent bias. 422 In the application of the disqualification standard, there should be congruency between the language of the text and the ordinary meaning of the words chosen to implement the text. In modern American usage, “might” is a word that occupies a place on the continuum of possibility. 423 The proper interpretative approach is one that analyzes the disqualification standard from the perspective of possibility, not probability. Reasonable suspicion is a metric that is congruent with the plain and interpretive meaning of the ethical mandate's “might.”

Adrian Vermeule provides prudent advice about interpretation in decision-making -- judges should stick close to the surface level or literal meaning of clear and specific texts, resolutely refusing to adjust those texts by reference to a judge's conception of textual purpose, drafters’ understanding, public values and norms. 424 In addition, consistent with Vermeule's advice, judicial implementation of the semantically clear disqualification standard should avoid unnecessary and potentially distorting adjectival amplifications of the evidential standard. The standard for the perception of judicial impartiality should not be qualified or amplified by terms like “substantial,” “significant,” or “serious,” which are often applied in an ad hoc fashion to the reasonable person's perception. 425 Similarly, application of the appearance standard should not be weakened or compromised by self-serving, balance-shifting procedural devices, such as presumptions. 426 If sufficient evidence is produced to undermine the presumption, the presumption should dissipate. 427

b. Contextual adjustment of the metric: Judicial impartiality is recognized as a value of the highest order, integral to the concept of a fair trial, a fair tribunal, and the public's confidence in our system of justice. 428 Accordingly, when such interests are implicated, the level of scrutiny should be adjusted to accommodate and protect those fundamental interests. 429 Fleming James suggests that in difficult cases, and to avoid a harsh or “unlovely” spectacle, courts may relax the requirements of proof. 430 In the specific context of apparent bias, the Shankar court stated:

The point simply is this: there is a vital public interest in subjecting the decisions of those engaged in any aspect of judicial or quasi-judicial work to the most exacting scrutiny in order to ensure that their decisions are not only beyond reproach in fact and indeed from the perspective of a lawyer or a judge but also beyond reproach from the perspective of a reasonable member of the public. The inquiry should be directed from the perspective at whether the events complained of provide a reasonable basis for such a person apprehending that the tribunal might have been biased. 431

Adjusting the level of scrutiny in accordance with the reasonable suspicion standard provides a sufficient baseline, as well as procedural flexibility, to protect the appearance of judicial impartiality in the difficult context of uncertainty, limited information, and the public's trust and confidence in the judicial system's integrity.

c. Minimizing the costs and risks of error: The recalibrated level of scrutiny (reasonable suspicion) provides protection from the harmful consequences of erroneous decision-making in disqualification cases. A recalibrated standard, faithful to the precept's text and values, would promote greater judicial caution in recusal matters and engender greater public confidence. Allocating the burden of uncertainty (especially when decision-making is dependent on the “objective” application of a vague metaphorical construct like the “reasonable” observer) is a challenging task. Vermeule has suggested various strategies, such as the maximin criterion and satisficing. In the former, some choices dominate others in the absence of probability information because the dominant choice produces better outcomes than the outcome of the alternative, and never produces a worse outcome. 432 In the latter, rather than adopting a maximizing strategy to pick the “best” option, one decides, in the face of constraints, to pick an option that is simply “good enough,” which can, as Vermeule notes, be a surprisingly good option for making accurate decisions. 433 The interesting aspect of these options is that the reasonable suspicion standard is an approach that serves a fundamental risk-averse principle that is often stated (but not sufficiently implemented) in disqualification cases – i.e., when in doubt, the jurist should err on the side of caution and disqualify. 434

d. The “reasonable” safety valve: In disqualification matters, judges seem to exhibit scorn for a claim that exemplifies “suspicion,” often cavalierly linking it with the adjective “mere.” 435 Sometimes, one senses that the real concern (misplaced) is with actual bias, often demonstrated by a defensive, good faith protestation of the jurist's unimpeachable impartiality. Australia decided to use different nomenclature and adopted a “reasonable apprehension” standard. 436 Whether one uses the terminology of apprehension or suspicion, the fundamental standard remains the same. Reasonable suspicion (or apprehension) is not mere suspicion -- it requires explanation and a careful articulation of the relevant facts and circumstances to support appearance-based recusal. Free-floating suspicion or unsupported belief will not, and should not, justify disqualification. Notwithstanding its semantically and psychologically slippery aspect, 437 “reasonable” is the indispensable anchor for principled decision-making in appearance-based disqualification.

e. Comparative jurisprudence: The discussion about the jurisprudence from the selected common law jurisdictions reflects a studious (and, at times, admittedly complicated) attempt to eventually reach a jurisprudential consensus in the quest for a prudent, principled, and practical standard governing apparent bias. In its application of the lay observer heuristic, common law countries have demonstrated a determination to protect cherished public values and promote public confidence. Whether the reasonable observer standard is considered in relation to “suspicion” or “apprehension,” the common law jurisdictions have gravitated toward a calibration that reflects a lower level of probability (viz., possibility). 438

f. Symbolic utility: Commentators have recognized the importance of the expressive aspect of a government's statements or actions. 439 Robert Nozick explains that the symbolic aspect of an action may sometimes be more important than a causal one and should be recognized as an important and independent factor in normative decision-making. 440 Ethics reflects the values we cherish and protect. Recusal decisions can attract public attention, especially if a case or jurist is high-profile. Just one instance of a controversial refusal to recuse can result in significant reputational (institutional and individual) harm. The loss of public trust and confidence is very difficult to repair or restore. Erring on the side of caution, based on clear ethical and jurisprudential principles, is the prudent course of action to maintain the public's trust and confidence in the rule of law.