AZC20 v Secretary, Department of Home Affairs

the issue 8 On 3 October 2023, the day before the hearing set down to deal with amendments to the statement of claim, my chambers received written submissions from the applicant seeking that I disqualify myself from hearing the case, together with an affidavit affirmed by the applicant's solicitor. The application was confirmed orally at the hearing on 4 October. In the light of this development, the hearing did not proceed. I gave directions for the respondents to file submissions in response and for a hearing of the application, if needed, on 13 October 2023. 9 My disqualification is sought in order to avoid what the applicant says is a reasonable apprehension of bias in the final determination of the hearing. The bases for that apprehension, briefly, are as follows. (a) The respondents rely on the evidence of Ms Davis to resist the argument that the duty to remove the applicant from Australia has not been pursued with efficiency and diligence. Broadly speaking, Ms Davis gives evidence as to the practicalities of removing detainees to Iran at various times including during the COVID-19 pandemic. Ms Davis gave evidence of a similar kind in three cases in which I appeared as senior counsel for the Commonwealth or one or more of its officers (Commonwealth parties). (b) Some reliance is also placed on my long association with the Commonwealth and its agencies, as counsel at the private bar and earlier as a government lawyer, although it appears to be accepted that this is not sufficient to give rise to a reasonable apprehension of bias in itself.

principles 10 The principles to be applied are not in contest. A reasonable apprehension of bias requires that there be a possibility, real and not remote, that a fair-minded and appropriately informed lay observer might reasonably apprehend that the decision-maker might not bring a fair, impartial and independent mind to the determination of the matter on its merits. Recently in QYFM v Minister for Immigration, Citizenship, Migrant Services and Multicultural Affairs & Anor [2023] HCA 15; 97 ALJR 419 at [38], Kiefel CJ and Gageler J identified the inquiry as involving the following steps: Application of the criterion was identified in Ebner, and has been reiterated, logically to entail: (1) identification of the factor which it is said might lead a judge to resolve the question other than on its legal and factual merits; (2) articulation of the logical connection between that factor and the apprehended deviation from deciding that question on its merits; and (3) assessment of the reasonableness of that apprehension from the perspective of a fair-minded lay observer. 11 It is important to note that "It is an open, and not an empty, mind that must be kept": SZQYM v Minister for Immigration, Citizenship, Migrant Services and Multicultural Affairs [2020] FCA 779; 169 ALD 579 at [102] (Allsop CJ). Thus, preliminary views - even fairly strong ones - about the merits of an issue do not found a reasonable apprehension of bias. To put it another way, an apprehension that the decision maker is likely to come to a certain conclusion is not an apprehension of bias in the relevant sense. What must be shown is a reasonable apprehension that the decision maker might not resolve the question on its merits. 12 A further point that should be noted at this stage is that judges should not accede too readily to suggestions of apprehended bias. A judge has a duty to hear cases that come before them and should generally not yield to the temptation of acceding to a request for recusal - thereby increasing the workload of another judge - to avoid awkwardness. To do so would be an abdication of duty and might encourage parties to make applications for recusal in an attempt to secure a judge who is thought to be more sympathetic: see, eg, Re JRL; Ex parte CJL (1986) 161 CLR 342, 352 (Mason J). (It was submitted by the applicant here that he is not engaged in "judge shopping". I express no view as to that, except that the motivations of the individual applicant are irrelevant to the point made by Mason J.)

Ms davis as a witness 13 The importance of Ms Davis's evidence in the present case is somewhat doubtful. However, it is not clear how the other evidence in the case will develop and I have not yet heard final submissions. I therefore proceed on the basis that the view that is taken of her credibility as a witness and her diligence as an officer may have an influence on the outcome. 14 Ms Davis has given evidence on similar issues in a number of other proceedings involving people in immigration detention. In three of those cases, I appeared as senior counsel for Commonwealth parties (the parties who relied on her evidence). 15 The first was AJL20 v Commonwealth [2020] FCA 1305; 279 FCR 549, which was heard in July 2020. Ms Davis swore three affidavits in that proceeding. She was called by junior counsel for the Commonwealth and cross-examined briefly by video link. Neither the transcript (which was in evidence on this application) nor the judgment indicates that there was any substantial challenge to the evidence that she gave. The basis upon which it is submitted that AJL20 was a "hard-fought" case is therefore unclear. At an evidentiary level, it was not: see at [95]-[115], [130]-[151]. 16 The second case was CYG20 v Commonwealth [2021] FCA 259 (CYG20). Ms Davis swore an affidavit in that case. She was not required for cross-examination and therefore was not called to give oral evidence. 17 The third case was Tran v Commonwealth [2021] FCA 580 (Tran). Ms Davis swore an affidavit in that case. The available evidence does not record (and I do not remember) whether she was required for cross-examination. In her reasons for decision, Jagot J said (at [35]): Some [of the Commonwealth's witnesses] were cross-examined, others not. There was no challenge to the credit of any of the witnesses. There was no real dispute about the facts, as the chronology discloses. The dispute focused on the inferences that should be drawn from the facts. 18 In the present case, my attention was drawn to AJL20 and I was asked, by way of an email to my chambers, to disclose the nature of my interactions with Ms Davis. I conducted a search on online databases, which brought to light CYG20 and Tran, and provided these references to the applicant's solicitors. I also informed them that I could not recall any further cases and that, so far as I could recall, I had never been in the same room as Ms Davis or conferred with her. The hypothetical fair-minded observer should be taken to know these things. 19 As noted recently in Masi-Haini v Minister for Home Affairs [2023] FCAFC 126 at [105] (Markovic, Meagher and Kennett JJ), the hypothetical fair-minded observer, despite being a lay person, should also be taken to know something of the legal system and barristers' work. Importantly, the observer is aware of the basic principle that a barrister is bound by their professional obligations to put, disinterestedly, the best case that can properly be put for a client. The case advanced by counsel for a client in contested proceedings is not to be taken to represent the personal views of counsel as to the merits of the case or the virtue (or otherwise) of the client or a witness. True it is, as the applicant submits, that counsel takes responsibility for selecting the witnesses to be called; and the hypothetical observer may therefore think it likely that counsel has made some assessment of a witness who is called. Certainly that is desirable, although in my experience it is not always able to be done. 20 However, I do not think it follows that counsel, having called a witness (or, as senior counsel, had their junior call a witness), would for that reason be regarded by a fair-minded observer as having a particular view (let alone a positive view) of that witness. The calling of a witness simply means that counsel: (a) is not aware that the witness's evidence is false (so that ethical duties do not prevent it being led); and (b) considers that the client is more likely to succeed with the witness's evidence than without it. Acceptance as to the latter point may depend on matters such as how likely the evidence is to be controversial and what other witnesses (if any) instructing solicitors have found or the client has made available. Reservations may be held about whether the witness will be, or even should be, believed. Further, that judgment is made only by reference to the particular case and the evidence the witness is likely to give in that case. It says little if anything as to how, having been appointed as a judge, the former barrister will view the witness's evidence in a later case. 21 The authorities accord with this view. 22 In Contract Mining Services Pty Ltd v Adelaide Brighton Cement Limited [2020] SASC 69 at [48]-[51] (Contract Mining), Livesey J rejected a suggestion of apprehended bias based on his prior association with a particular witness. His Honour had called the witness in an earlier proceeding in which he was senior counsel, around a year before his appointment, and submitted that the witness's evidence was (inter alia) "impressive". Livesey J disqualified himself for other reasons, arising from having represented the defendant's interests in proceedings involving similar issues that spanned two years and ended the day before his appointment. 23 Gascor v Ellicott [1997] 1 VR 332 (Gascor) concerned the converse situation but is nevertheless instructive. It involved an application to remove an arbitrator. When practising as a barrister, the arbitrator had appeared in an earlier arbitration involving similar issues, vigorously cross-examined some of the witnesses who were to be called in the instant arbitration and criticised their evidence in his submissions. It was said that the arbitrator might thereby appear to have formed views about the reliability or credibility of those witnesses. Ormiston JA said at 349, 351-352: A cautious approach should be taken, according to the authorities, as to the degree of sophistication and knowledge of the reasonable observer for the purposes of the now accepted rule as to bias but I cannot treat such an observer as so ignorant as not to understand the well-recognised function of counsel briefed in proceedings for one party or the other. It would be both unreasonable and unworldly to ascribe to the fair-minded observer the degree of ignorance necessary to lead to the disqualification for bias of a judge or arbitrator whose only sin was that he or she had in the past vigorously attacked a witness in cross-examination and thereafter in submissions. … Throughout their submissions the arguments by counsel on behalf of the appellant rested upon a reasonable apprehension that Mr. Ellicott had "formed views about the relevant issues or common witnesses". The High Court has said nothing about the forming of views, whatever that might mean, but has directed attention to a reasonable apprehension that a judge or arbitrator may have prejudged an issue contrary to the expected contentions of a party in a later dispute. If a judge or arbitrator has formally expressed conclusions in written or oral reasons which may reasonably be perceived as unfavourable to a party and of a kind where it might be apprehended that the judge or arbitrator might find it difficult subsequently to reach a contrary conclusion, then the test for disqualification may well have been satisfied. It is possible also that an opinion may otherwise have been expressed either in a formal document or in such deliberate and strong terms as to lead to a reasonable apprehension of prejudgment: cf. Laws v Australian Broadcasting Tribunal (1990) 170 CLR 70. But to say that an arbitrator should be disqualified for bias because as barrister he may have "formed views" about some person or issue is to turn the test on its head. Unless a judge, arbitrator or tribunal member has gone into some form of monastic seclusion it would be impossible for such members of a judicial or quasi-judicial tribunal not to have formed views about a wide variety of issues and about people whom those members may have come across in the course of their professional activities. Those views may have been formed for one purpose or another and may be formed after consideration of a little or, it may be, a great deal of material, but they rarely result from a considered evaluation of all the evidence and arguments from both sides of a dispute. It is possible that after a long and complex arbitration a person appearing in it will have formed some opinion as to the reliability of a particular person as a witness but, unless it has been necessary to form an opinion for the purpose of advising say on an appeal, a barrister is unlikely to have done more than form a fleeting impression before passing to the next brief. No doubt from time to time a barrister will have formed a view that one expert witness is better than another, but that will be for practical forensic purposes as counsel must do the best they can for their clients with the witnesses available to be called. The question at issue is always possible prejudgment and, although the onus of proof as to a reasonable apprehension of the relevant kind appears to be relatively slight according to High Court authority, what is required is a reasonable apprehension as to the possibility that the arbitrator has formed a considered opinion of a kind which might lead to preconceptions on any relevant issue. This is not such a case. What I have just said does not gainsay the possibility that a judge or arbitrator or other tribunal member may have formed such a strong opinion of the untrustworthiness or unreliability of a particular proposed witness, even an expert witness, that such a tribunal member should disqualify him or herself from hearing a dispute. Nor would I exclude the possibility that the evidence will exhibit such a clear antagonism to or distrust of a particular witness, whether appearing from the form or nature of cross-examination, intemperate description in submissions, or from other source altogether, that the requisite apprehension of possible bias might be made out. This, again, is not such a case. 24 Tadgell JA agreed with Ormiston JA on this issue (at 344) and Brooking JA agreed with Tadgell JA. 25 Both of these cases involved more intensive prior involvement with witnesses and their evidence than I have had with Ms Davis. Although Gascor involved an arbitrator rather than a judge, there is no material difference in the principles that apply in relation to apprehensions of bias. If anything, an arbitrator should arguably be more readily disqualified than a judge because the considerations mentioned at [12] above do not apply. An arbitrator has no duty to accept appointment, and parties to an arbitration commonly have a choice as to who to appoint. 26 In submissions in reply, the applicant submits that cases decided before QYFM must be treated with caution because of what is said to be the changing test for apprehended bias and the changing perceptions of fair-minded lay observers. The suggestion that the nature of the test is changing is perplexing: all members of the Court in QYFM treated the principles as having been settled since Ebner v Official Trustee in Bankruptcy [2000] HCA 63; 205 CLR 337 if not earlier: see at [26], [37] (Kiefel CJ and Gageler J), [67] (Gordon J), [162] (Edelman J), [194] (Steward J, referring to CNY17 v Minister for Immigration and Border Protection [2019] HCA 50; 268 CLR 76 at [57]), [225] (Gleeson J), [273]-[274] (Jagot J). I do not regard anything said in the cases referred to above as inconsistent with the double "might" test established in Ebner. However, there are suggestions in the reasons in QYFM that the application of the test may vary from time to time according to understandings of the characteristics to be attributed to the hypothetical fair-minded observer, because the observer needs to be placed in a contemporary setting: at [46], [54] (Kiefel CJ and Gageler J), [83] (Gordon J), [275] (Jagot J). Three things should be said about this. 27 First, Gordon J observed at [83] that three intermediate appellate decisions should be considered "wrong by modern standards". Her Honour did not indicate whether this meant that those decisions were wrong when decided (between 1998 and 2008) or that more recent developments had made their reasoning inapplicable. Kiefel CJ and Gageler J, at [53], referred to the same decisions, expressly left open whether they were correct when decided, and concluded that they should not now be followed. This points to a deeper question as to what the Court is doing in concluding that the understandings of the hypothetical fair-minded observer, by reference to whom the existence of a reasonable apprehension of bias is found to exist or not, develop over time. 28 The hypothetical fair-minded observer is both hypothetical (ie not a real person) and fair-minded (a description which connects directly with the concept of a reasonable apprehension of bias). Their views and understandings are not established by calling evidence in particular cases about what ordinary people think, or by referring to statistics or scholarly works on the subject. My understanding, therefore, is that the hypothetical fair-minded observer is a construct called into existence in order to assist in giving content to the concept of a reasonable apprehension; that is, to assist in applying the test of reasonableness. Fair-mindedness is thus at least to some degree normative, rather than a question of ascertaining community attitudes. This is not to deny that what the hypothetical fair-minded observer thinks or understands can change over time. However, it means that the change is a function of the development of the common law rules of statutory construction (in the sense discussed in Plaintiff S10/2011 v Minister for Immigration and Citizenship [2012] HCA 31; 246 CLR 636 at [97] (Gummow, Hayne, Crennan and Bell JJ)). What is changing here is the courts' understanding of what apprehensions of bias are reasonable. 29 That means that earlier decisions which no longer accord with contemporary standards are being overruled as the common law (or, in traditionalist terms, understanding of the common law) changes. It is not the role of a single judge to overrule decisions of intermediate appellate courts (or to decide that another single judge is plainly wrong) on the footing that the law has changed. Even if what I have just said is not correct as a matter of technical analysis, a single judge should be slow to conclude that decisions from a few years ago, or even two or three decades ago, no longer represent the law, unless they clearly cannot stand with the reasoning in later High Court decisions. 30 Secondly, the specific change in the understandings of the hypothetical fair-minded observer noted in QYFM was a lessening in deference to authority and a corresponding appreciation that decision-makers, including judges, can be affected by human frailty (at [297] (Jagot J)). In the light of that observation, it is likely that less credence than formerly should be given to suggestions that judges (in the view of the fair-minded observer) are fitted by their training and required by their judicial oath or affirmation to administer justice dispassionately and ignore irrelevant matters. However, it does not follow that there is any reason to depart from the position that the hypothetical fair-minded observer understands essential facts about the operation of the legal system and (relevantly here) the work of advocates. 31 Thirdly, QYFM concerned the position of a judge of this Court (Bromwich J) who, in his previous role as the Commonwealth Director of Public Prosecutions (DPP), had argued points of law in the appellant's appeal against his conviction for drug importation offences. That conviction was the occasion for the cancellation of the appellant's visa, which was the subject matter of the proceeding that came before a Full Court including his Honour. Existing authority supported the view that a reasonable apprehension of bias did not arise from Bromwich J's earlier role in the prosecution of the appellant. That authority was reconsidered in the appeal to the High Court. 32 The position of prosecuting counsel in respect of an accused is different from that of other advocates in criminal, and certainly in civil, proceedings. The prosecutor - especially if they occupy a senior position in the prosecuting agency - will have formed the view that there was an arguable case that the accused was involved in criminal activity and then devoted effort to making good that case. The prosecutor is expected to act as a disinterested minister of justice, but at the same time acts on their own opinion that the prosecution of the accused is appropriate. It is easy to see why a fair-minded lay observer might entertain a fear that the former prosecutor might retain an impression of the accused that would affect their ability later, as a judge, to decide a case involving that person according to its merits. In QYFM there was the additional factor that the proceedings which came before the Full Court had some connection with the conviction that Bromwich J, as DPP, had successfully defended in the earlier appeal (see at [55] (Kiefel CJ and Gageler J), [84] (Gordon J)). 33 This situation is markedly different from that of a judge hearing evidence from a witness who gave evidence in an earlier case involving the judge as counsel. No doubt, in both situations, the particular facts of the individual case are important. Just as a judge who formerly prosecuted a person now before the court is not per se disqualified (QYFM at [82] (Gordon J)), it may be that a judge who formerly called a particular witness many times and defended their evidence might conceivably be disqualified from a case in which that witness's credit will be in issue. However, that is not the situation here. 34 Two other points should be mentioned at this stage. 35 First, the cross-examination of Ms Davis in the present case proceeded in an unusual way. She gave evidence by video link, apparently from her own office. At one point, cross-examining counsel asked her to log in to her work computer and search for some information unrelated to her evidence. At another point, when she was having some trouble with the terminology of a question she had been asked, she used her smartphone to conduct an internet search for the term (a fact which emerged when counsel asked her what she was looking at). This was clearly irregular. It was observed in the applicant's submissions that I gave no direction in relation to Ms Davis's use of the internet during her oral evidence. The nature of the direction that should have been given was not specified. No party asked for any direction to be made. These curious events may ultimately affect the assessment of Ms Davis's evidence (which will be a matter for submissions). I do not think that they add anything to any concern that a hypothetical observer might have concerning my capacity to bring an open mind to that assessment. 36 Secondly, it also emerged during Ms Davis's cross-examination that she was exchanging text messages with another officer, Mr Paul Wyllie. This clearly should not have happened, at least without notice to the Court and the applicant's legal team. It is another problematic aspect of Ms Davis's oral evidence, which is likely to affect the assessment of her credibility if her evidence is controversial. 37 It is noted by the applicant that Mr Wyllie also gave evidence in a case in which I appeared as counsel for the Minister. The case was Plaintiff S195/2016 v Minister for Immigration and Border Protection [2017] HCATrans 25, an application for urgent interlocutory relief in which Mr Wyllie was not cross-examined. It is possible that I appeared in other matters in which Mr Wyllie gave evidence, but neither my research nor that of the applicant's solicitors has uncovered them. 38 It is submitted by the applicant that this raises a further problem because Mr Wyllie is also a witness in the case, and my answer to the applicant's inquiry concerning my dealings with witnesses was therefore incomplete. Other than in submissions on this disqualification application, I have not been referred to an affidavit by Mr Wyllie or informed that he is to be cross-examined. I do however accept that, pursuant to an order I made by consent on 26 July 2023, an affidavit Mr Wyllie affirmed on 11 May 2021 in VID89/2021 (the proceeding in which Rangiah J delivered AZC20 FC) is to be taken as an affidavit in these proceedings. My response to the applicant's inquiry (which referred to witnesses in the proceeding) was therefore incomplete, as attention was focused at that time on Ms Davis. However, I do not think that adds anything to the present issue. As a matter of substance, my dealings with Mr Wyllie as a witness (so far as the evidence shows, and so far as I can recall) have been even less consequential than those with Ms Davis.

association with commonwealth parties 39 The applicant's submissions also draw attention to my long association, as a lawyer, with Commonwealth parties. (a) The affidavit of the applicant's solicitor draws on a publicly available source to assert that I worked as a solicitor for the Australian Government Solicitor before commencing practice at the private bar. This is not correct. However, it is true (and would be known to a hypothetical fair-minded observer) that for several years in the 1990s I was a government lawyer in the Commonwealth Attorney-General's Department. (b) The applicant's submissions then present a summary of my appearances as a barrister in matters arising under the Act. The source referred to for this summary is the solicitor's affidavit, but no such material appears in the affidavit. The last migration case referred to in which I appeared against Commonwealth parties is a 2008 matter in the (then) Federal Circuit Court; however, this ignores (at least) Guo v Commonwealth [2017] FCA 1355; 278 FCR 31. I therefore do not have much confidence in the detailed figures presented. (c) However, the overall point is correct. As counsel I appeared in a large number of cases under the Act; and in nearly all of these cases I appeared for Commonwealth parties. I began doing this work early in my practice; and the last case in which I appeared (in March 2022, a few weeks before my appointment to the Supreme Court of the Australian Capital Territory) was a migration matter in which I appeared for the Minister. All of this is available from the public record to anyone who is sufficiently interested to make the relevant internet searches. I proceed on the basis that it is known to the hypothetical fair-minded observer, even though it has not been properly established by evidence. 40 The applicant accepts that these matters in themselves are not sufficient to give rise to a reasonable apprehension of bias. This is correct (see, eg, British American Tobacco Australia Ltd v Gordon [2007] NSWSC 109 at [85] (Brereton J)). They are said to "give colour" to the other facts and render the issue that they raise "more acute". I am not sure exactly what this means. I apprehend it to mean that the hypothetical fair-minded observer, when considering what inferences might arise from the facts concerning Ms Davis (discussed above), would have in mind that I appeared as counsel for Commonwealth parties in a very large number of migration cases and for opposing parties in very few. 41 I do not think there is any reason in principle why a reasonable apprehension of bias cannot arise from two or more factors working together. However, such circumstances must be rare and they do not exist here. I will endeavour to explain why this is so. 42 It has often been remarked that judges and other decision-makers inevitably have opinions about a wide range of matters. Thus, it is common, if not usual, for a judge to approach a case with a preliminary view as to what is likely to be the correct outcome. Such leanings do not constitute bias, and the apprehension that the judge has a preliminary view (even a strong one) is not a basis for disqualification. So, for example, in ALA15 v Minister for Immigration and Border Protection [2016] FCAFC 30, analysis showing that a judge had decided more than 99 percent of Migration matters adversely to applicants, mostly summarily, was held to be irrelevant to establishing an apprehension of bias (see especially at [39]). The essence of bias is the inability to analyse evidence and argument on their merits and to be persuaded, in an appropriate case, that one's preliminary view was wrong. An apprehension of bias therefore necessarily involves pointing to a factor (or possibly a combination of factors) that might reasonably be thought to cause that resistance to persuasion. It is not enough to demonstrate an apprehension that the decision-maker is probably going to decide the case one way rather than another. 43 It is for this reason that apprehended bias cases usually if not always focus on particular issues that are said to have the potential to compromise the decision-maker's ability to decide a case on its merits: eg prejudgment (arising from having previously stated a strong view on an issue or decided the same issue against a party); or association (familial or social) with a party, witness or advocate. Such factors, if made out, indicate more than a leaning towards a result; they indicate a potential impediment to a decision on the merits. In most cases, therefore, what was referred to in QYFM at [38] as "the factor which it is said might lead a judge to resolve the question other than on its legal and factual merits" and "the logical connection between that factor and the apprehended deviation from deciding that question on its merits" will involve a particular aspect of the decision maker's background or conduct and a particular danger of deviation from proper decision making. 44 In every case, the "factor" that is relied upon must be considered in its whole context. However, the context is relevant to the extent that it accentuates (or diminishes) the possible effect of that factor and not otherwise. Adding together two substantially discrete factors, each said to be capable of pointing the decision maker in a particular direction, does no more than establish a leaning in that direction. 45 In the present case, the fact that I have appeared in cases where my clients relied on the evidence of Ms Davis and Mr Wyllie and the fact that I nearly always appeared for Commonwealth parties in migration cases are not factually unconnected, but they are analytically distinct. To the extent that the former factor raises a concern, it would raise the same concern if the only migration cases I had appeared in were those involving Ms Davis. Nothing is added by the latter factor, in light of the acceptance that it does not give rise to an apprehension of bias. 46 In any event, for the reasons explained above, I am firmly of the view that no reasonable apprehension of bias arises from the cases in which I appeared and in which Ms Davis gave evidence. It is not a borderline case that might be pushed across the line by the "colour" provided by another factor.

Disposition 47 The disqualification application is refused. 48 I will list the matter for case management on a date to be fixed, to determine whether the existing timetable can be adhered to and what ought to be done to achieve that. At that time I will also hear any submissions that the parties wish to make concerning the costs of this application. I certify that the preceding forty-eight (48) numbered paragraphs are a true copy of the Reasons for Judgment of the Honourable Justice Kennett.