R v Tarrant

Principles applicable to disqualification for apprehended bias

1. The principles applicable to a recusal or disqualification application upon the ground of apprehended bias are stated in Ebner v The Official Trustee in Bankruptcy [2000] HCA 63; (2000) 205 CLR 337 at 344 - 345 by the authors of the joint judgment as follows (citations omitted): "[6] Where, in the absence of any suggestion of actual bias, a question arises as to the independence or impartiality of a judge (or other judicial officer or juror), as here, the governing principle is that, subject to qualifications relating to waiver (which is not presently relevant) or necessity (which may be relevant to the second appeal), a judge is disqualified if a fair-minded lay observer might reasonably apprehend that the judge might not bring an impartial mind to the resolution of the question the judge is required to decide. That principle gives effect to the requirement that justice should both be done and be seen to be done, a requirement which reflects the fundamental importance of the principle that the tribunal be independent and impartial. It is convenient to refer to it as the apprehension of bias principle. [7] The apprehension of bias principle may be thought to find its justification in the importance of the basic principle, that the tribunal be independent and impartial. So important is the principle that even the appearance of departure from it is prohibited lest the integrity of the judicial system be undermined. There are, however, some other aspects of the apprehension of bias principle which should be recognised. Deciding whether a judicial officer (or juror) might not bring an impartial mind to the resolution of a question that has not been determined requires no prediction about how the judge or juror will in fact approach the matter. The question is one of possibility (real and not remote), not probability… [8] The apprehension of bias principle admits of the possibility of human frailty. Its application is as diverse as human frailty. Its application requires two steps. First, it requires the identification of what it is said might lead a judge (or juror) to decide a case other than on its legal and factual merits. The second step is no less important. There must be an articulation of the logical connection between the matter and the feared deviation from the course of deciding the case on its merits… ."
2. In Johnson v Johnson [2000] HCA 48; (2000) 201 CLR 488 at 492 - 493 the following statements appear in a judgment joined in by five of the justices (citations omitted): "[11] It is not contended that Anderson J was affected by actual bias. It has been established by a series of decisions of this Court that the test to be applied in Australia in determining whether a judge is disqualified by reason of the appearance of bias (which, in the present case, was said to take the form of prejudgment) is whether a fair-minded lay observer might reasonably apprehend that the judge might not bring an impartial and unprejudiced mind to the resolution of the question the judge is required to decide. [12] That test has been adopted, in preference to a differently expressed test that has been applied in England, for the reason that it gives due recognition to the fundamental principle that justice must both be done, and be seen to be done. It is based upon the need for public confidence in the administration of justice. 'If fair-minded people reasonably apprehend or suspect that the tribunal has prejudged the case, they cannot have confidence in the decision'. The hypothetical reasonable observer of the judge's conduct is postulated in order to emphasise that the test is objective, is founded in the need for public confidence in the judiciary, and is not based purely upon the assessment by some judges of the capacity or performance of their colleagues. At the same time, two things need to be remembered: the observer is taken to be reasonable; and the person being observed is 'a professional judge whose training, tradition and oath or affirmation require [the judge] to discard the irrelevant, the immaterial and the prejudicial'."
3. In Webb v The Queen [1994] HCA 30; (1994) 181 CLR 41 at 74 Deane J listed the different categories of circumstances which might give rise to an apprehension of bias as follows (citations omitted and emphasis added): "The area covered by the doctrine of disqualification by reason of the appearance of bias encompasses at least four distinct, though sometimes overlapping, main categories of case. The first is disqualification by interest, that is to say, cases where some direct or indirect interest in the proceedings, whether pecuniary or otherwise, gives rise to a reasonable apprehension of prejudice, partiality or prejudgment. The second is disqualification by conduct, including published statements. That category consists of cases in which conduct, either in the course of, or outside, the proceedings, gives rise to such an apprehension of bias. The third category is disqualification by association. It will often overlap the first and consists of cases where the apprehension of prejudgment or other bias results from some direct or indirect relationship, experience or contact with a person or persons interested in, or otherwise involved in, the proceedings. The fourth is disqualification by extraneous information. It will commonly overlap the third and consists of cases where knowledge of some prejudicial but inadmissible fact or circumstance gives rise to the apprehension of bias."
4. This may not be an exhaustive or exclusive list of the categories, as was observed by the authors of the joint judgment in Ebner v The Official Trustee in Bankruptcy at [23]. The present application is clearly in the second category. It is said that my conduct in the following respects would give rise to a reasonable apprehension of bias in a fair minded lay observer: 1. During the trial of Roff I asked questions of Tarrant in the course of her evidence in the Crown case. 2. During the trial of Roff I did not rebuke defence counsel in the presence of the jury for the use of coarse language in two questions which he asked of Tarrant. 3. During the trial of Roff I did not rebuke defence counsel in the presence of the jury for questioning Tarrant concerning her sexual relationship with the deceased. 4. During Tarrant's own trial I asked questions of her concerning her apprehension of violence at the hands of Alois Rez in the course of their de facto marital relationship. 5. In proceedings after conviction in relation to Roff I made observations concerning the state of evidence and issues to be resolved in the sentence proceedings relating to both Roff and Tarrant.

The Roff trial - general considerations re trial judge's questions of Tarrant

1. Set out under a series of headings below are my reasons for rejecting the submissions now made on behalf of Tarrant that some of my interventions in the questioning of her during the trial of Roff have given rise to a reasonable apprehension of bias against her with respect to sentencing her for manslaughter. These headings address the context and purpose of each of the several impugned passages of trial judge's questions respectively. There is a consideration which negates generally any apprehension of bias arising from such questioning, irrespective of the content as considered passage by passage under the subsequent headings. Namely that the extent to which it was proper for me as trial judge to question Tarrant as a Crown witness in the Roff trial was dictated by established principles, the content of which would make clear to any informed reasonable observer that I was required in the Roff trial to judge what interventions in questioning might be necessary or least appropriate to ensure that the jury gained a proper understanding of her evidence, including by seeking clarification of her answers and obtaining her response to perceived or potential improbabilities or contradictions. At the same time I was required to avoid the appearance of endorsing or rejecting either party's case.
2. As trial judge I was not required to take into account any sensitivity or delicacy of feeling of the witness nor to ask questions or to refrain from doing so for the purpose of maintaining impartiality towards her as a witness. It makes little sense to speak of impartiality toward a witness as no judgment or ruling is required to be made affecting the witness. The overriding consideration is the maintenance of impartiality, in fact and in appearance, toward the accused and the Crown. This is apparent from the statements of principle extracted from the following authorities. If it were otherwise the consideration of trying to preserve an impression of impartiality towards a witness would in many cases - and the trial of Roff is one of them - run counter to and interfere with doing justice to either the Crown or the accused or both in the trial.
3. Knowing of these constraints a reasonable observer would take account of the different capacities of Sarah Tarrant in separate proceedings. Namely, on the one hand as a witness in the trial of Roff wherein the trial judge in fairness to the parties in that trial might feel obliged to question her to certain effect and, on the other hand, as a convicted person standing before the same judge for sentence, at which time the judge could be expected to deal with her for her own crime impartially, according to law and the evidence.
4. In Galea v Galea (1990) 19 NSWLR 263 at 281 Kirby ACJ extracted from the cases the following principles concerning complaints of excessive judicial intervention (reproduced only so far as potentially relevant to the present case and omitting altogether his Honour's item 6 which was concerned only with civil cases). These principles are expressed in terms of when appellate intervention may be justified on account of excessive or improper judicial questioning of witnesses. However they serve to show that the constraints upon judicial questioning are regulated by the consideration of whether a fair hearing has been afforded to the parties before the court, not by regard for maintaining an appearance of impartiality towards witnesses: "1. The test to be applied is whether the excessive judicial questioning or pejorative comments have created a real danger that the trial was unfair. If so, the judgment must be set aside: see E H Cochrane Ltd v Ministry of Transport [1987] 1 NZLR 146. 2. A distinction is drawn between the limits of questioning or comments by a judge when sitting with a jury and when sitting alone in a civil trial. Although there is no relevant distinction, in principle, between the judicial obligation to ensure a fair trial whatever the constitution of the court, greater latitude in questioning and comment will be accepted where a judge is sitting alone. This is because it is conventionally inferred that a trained judicial officer, who has to find the facts himself or herself, will be more readily able to correct and allow for preliminary opinions formed before the final decision is reached: see R v Matthews (1983) 78 Cr App R 23; E H Cochrane Ltd v Ministry of Transport. 3. Where a complaint is made of excessive questioning or inappropriate comment, the appellate court must consider whether such interventions indicate that a fair trial has been denied to a litigant because the judge has closed his or her mind to further persuasion, moved into counsel's shoes and 'into the perils of self persuasion': see Sir Robert Megarry 'Temptations of the Bench' (1978) 16 Alta L Rev 406 at 409; see also U Gautier, 'Judicial Discretion to Intervene in the Course of the Trial' (1980) 23 Crim LQ 88 at 95-96 and cases there cited. 4. The decision on whether the point of unfairness has been reached must be made in the context of the whole trial and in the light of the number, length, terms and circumstances of the interventions. It is important to draw a distinction between intervention which suggests that an opinion has been finally reached which could not be altered by further evidence or argument and one which is provisional, put forward to test the evidence and to invite further persuasion: see In Marriage of Lonard (1976) 26 FLR 1 at 10-11; 11 ALR 618 at 626; see discussion (1976) ACLD DT 630; cf Ex parte Prentice; Re Hornby (1969) 90 WN (NSW) (Pt 1) 427; [1970] 1 NSWR 654. 5. It is also relevant to consider the point at which the judicial interventions complained of occur. A vigorous interruption early in the trial or in the examination of a witness may be less readily excused than one at a later stage where it is designed for the legitimate object referred to in Jones, namely of permitting the judge to better comprehend the issues and to weigh the evidence of the witness concerned. By the same token, the judge does not know what is in counsel's brief and the strength of cross-examination may be destroyed if a judge, in a desire to get to what seems crucial, at any stage prematurely intervenes by putting questions: seeYuill (at 185) and Gautier (at 117)."
5. In R v Esposito (1998) 45 NSWLR 442 at 472 Wood CJ at CL stated the principles as follows, specifically with respect to the position of the presiding judge in a criminal trial: "The line that a trial judge walks when asking questions of a witness is a narrow one. There is nothing wrong with questions designed to clear up answers that may be equivocal or uncertain, or, within reason, to identify matters that may be of concern to himself. However, once the judge resorts to extensive questioning, particularly of the kind that amounts to cross-examination in a criminal trial before a jury, then he is treading on thin ice. The thinness of that ice will depend upon the identity of the witness being examined (here the person on trial), and on whether the questions appear to be directed towards elucidating an area of evidence that has been overlooked or left in an uncertain or equivocal state, or directed towards establishing a point that is favourable or adverse to the interests of one or other of the parties."
6. In that case the trial judge's extensive questioning of the accused herself was held to have denied the accused fair trial and resulted in a conviction for murder being quashed. The extent and nature of the questioning was in stark contrast to the minimal intervention by myself in the trial of Roff, in questioning a Crown witness. The case provides a counter example of a degree of questioning from the bench which might, within the context of the trial itself, indicate an inclination on the part of the trial judge against one of the parties.
7. For what relevance it may have, the examination in chief of Tarrant in the trial of Roff occupied one half days over 118 pages of transcript. Cross-examination occupied about the same amount of time and filled about the same number of pages of transcript as the examination in chief. No complaint is now made that my own questions, overall, were excessive in number; only that a few passages are said to give rise to an apprehension of bias in relation to the sentencing of Tarrant for manslaughter.
8. I was not at any time during the Roff trial asked by either counsel to refrain from asking questions or to direct the jury to disregard any answers I elicited. During summing up I provided to the jury a document marked SU2 which contained general directions about how they should approach making use of the evidence they had heard and concerning the distinct functions of the judge and jury. This included a direction, in section 1, that the they "are the sole judges of the facts" and should "disregard any view apparently held by the trial judge unless this accords with your own". When reading this written direction over to the jury at p 5 of the summing up I said: "I have asked questions of witnesses. If you think the answers were of any use to you, you act on them. But you are not to be influenced by the fact that I intruded to ask questions."
9. Neither counsel sought any redirection in this respect. Clearly both accepted that the extent of questioning from the bench during the evidence of Tarrant was acceptable in accordance with the principles from the cases cited above. As the questioning of her was in accord with the requirements of a fair trial of Roff it could not reasonably be taken by any properly informed observer as having any bearing at all upon the impartiality with which I could be expected to deal with Tarrant in the entirely distinct setting of sentence proceedings in relation to her own conviction.

The Roff trial - questions on Tarrant's love letter

1. The first selection from the questions I asked Tarrant in the trial of Roff which is now said to give rise to an apprehension of bias is the sequence which appears at T 433.31 - 434.6. This concerned a love letter which Tarrant said she had written to Roff in March 2013. It was in evidence at Tab 8 of Ex C in the trial.
2. Prior to my questions (at the transcript reference given above), the Crown commenced to question Tarrant about the love letter from T 431.1. Over the following two and a half pages the Crown asked her to elaborate the meaning of some of the content of the letter.
3. The following was part of the letter: "There was only one thing that I regret that I did, that was taking the morning after pill so I wouldn't become pregnant, knowing that I had gone against everything that I have said and promised to myself that I would never do and I did. Getting rid of our baby before it ever had time to become a baby it destroyed me tremendously, you entered into a I wouldn't even know what to call it, you just lost it took off and I never heard from you for four hours until I came and found you. You scared me that night I thought I was going to loose (sic) you to which I wasn't ready for not after I had given up a baby that cold (sic) have been our own. I was so distraught and I still am now. That was just one of the hurdles we have jumped that made us stronger and closer together than ever before."
4. At T 433.11 - .25 Tarrant's attention was directed by the Crown to this passage and she was asked whether what she had written with respect to "taking the morning after pill" expressed "how you felt about that". It appeared to me that the passage quoted above from the letter was uncertain as to what was intended by the words "everything that I have said and promised to myself that I would never do and I did". The questions I asked which are now said to indicate a bias against Tarrant sought further explanation in the context that at this point the Crown was eliciting, generally, what she had meant by the document.
5. It is submitted that my questions "suggested" hypocrisy on the part of Tarrant in that she answered to the effect she had a strong moral conviction against causing the death of a foetus whereas she was admitting in the balance of her evidence that she had planned the killing of Alois Rez. I do not consider that a reasonable observer would consider that this or any other suggestion was being made by the trial judge. The questions were open, non-leading and did not "suggest" anything from the questioner. Further, if her answers might be characterised as exhibiting hypocrisy I do not see that that of itself could be of any significance in relation to her sentence, which is to be imposed for her admitted role in killing Alois Rez pursuant to a plan formed with her lover.
6. It is further submitted that "Tarrant's personal beliefs as to abortion and the 'morning after' pill were completely irrelevant to the issues the jury needed to determine in Roff's trial". That was a matter for the Crown and Roff's defence counsel to determine. The Crown had seen fit to tender the letter which conveyed, albeit not with complete clarity, Tarrant's sentiments concerning her use of the "morning after pill". The defence had taken no objection to the document and had not requested that this part of it be excised. Asking questions to clarify the meaning of a piece of writing which was in evidence, following the Crown already having done this to some extent, did not involve the introduction by the trial judge of any subject that was not, in any event, before the jury.
7. It is also submitted on behalf of Tarrant that the answers elicited by my questions "had the potential to invite divisive, irrelevant discussion amongst the jurors (about the rights and wrongs of pregnancy termination)". Whether or not that was so was and remains none of Tarrant's concern. It was a matter entirely for the consideration of the Crown and of Roff in the running of his trial. If either of the parties had had a concern in this respect it was open to either of them to seek a specific direction to the jury to avert "divisive, irrelevant discussion". This alleged possibility could not reasonably generate an apprehension of judicial bias against Tarrant in relation to her sentence.
8. In par 14 of the submissions in support of recusal it is argued that my questions on the subject were asked "only for the purposes of making an adverse finding against Tarrant in her own sentencing proceedings". That is a bare and unfounded assertion and it is rejected. The questions were asked for the purpose of clarification as explained at [25]. It is further said in this paragraph of the submissions that the questions were asked "unfairly, in that it was in the absence of Tarrant's legal representatives". There is no obligation on a trial judge to ensure the attendance of legal representatives for a witness or to afford such representatives opportunity to object or to make submissions in order to protect the witness against having to give answers which might appear to bear upon an assessment of the witness for some other purpose in some other proceeding. The proper concern of the presiding judge in the trial of Roff was to ensure fairness to him and to the Crown. As indicated by the lack of objection or request for specific instruction in the summing up, the parties at the trial evidently found no unfairness to them arising from the questioning.

The Roff trial - questions on the events of 23 July 2013

1. Tarrant's evidence in the trial of Roff included an account of an altercation between herself and Alois Rez on 23 July 2013. Two police officers called in the Crown case also gave evidence about these events. Roff in due course gave evidence about them himself. The combined effect of the evidence was as follows. During the evening of 23 July 2013 Roff visited 46 Alfred Street and remained there for some hours. He left at about 10:30 pm. After his departure Alois Rez accused Tarrant of conducting an affair with Roff. There was a physical altercation between them. Tarrant sent a text message to Roff to the effect that Alois Rez had tried to drag her by the hair down the hallway of the house and to eject her. She asked Roff to call the police. One of her text messages at this time also informed Roff that Alois Rez had said during the late evening that Roff was not to visit the house any more.
2. Roff called the police on the emergency services number. When they attended 46 Alfred Street Tarrant told them there was no problem. This caused the police to attend upon Roff himself to enquire what had prompted his emergency call. The episode angered Roff against Tarrant. He told her this had been their chance to have the police remove Alois Rez and that they would not have another such opportunity. Instead of capitalising on the chance she had made a fool of him.
3. I asked Tarrant questions about one aspect of these events at T 461.8 - .32. Those questions have now given rise to a submission that they would lead to a reasonable apprehension of bias on my part against Tarrant with respect to sentencing her. The aspect to which the impugned questions were directed was why Tarrant had not called Roff to retract her request that he summon the police after the altercation between herself and Alois Rez had subsided, given that the house had been peaceful for two hours prior to the police attendance.
4. In the passage of questioning by the Crown from T 460.4 - 461.6, immediately before my own questions, Tarrant had said that "it was a couple of hours" between her first asking Roff to call the police and the police arriving. She said in that interval Alois Rez "went into the bedroom and I went into the kitchen and started cleaning up and doing the dishes. I think I eventually went into the lounge room and he threw a doona at me so I had a blanket to sleep with".
5. Tarrant was the single most important witness in the Crown case. It was apparent from the defence opening at the outset of the trial that her credibility would be in issue. The Crown had opened that she would give evidence of the preconcert between herself and Roff to stupefy Alois Rez with sleeping tablets and then kill him. It was opened for Roff that he made no such agreement, that he did not kill Alois Rez and that he did not even know whether Alois Rez was dead. Given she was a witness whose credibility was to be attacked concerning events central to the Crown case, it was relevant for Tarrant to be questioned as to an apparent anomaly in her narrative. Such questions could have come from the Crown as her evidence in chief unfolded, or later from defence counsel. The only relevant restraint upon the trial judge asking such questions was that the manner and extent of questioning should not give to the jury the appearance of the judge taking one side. Neither party objected to this particular anomaly being addressed by questions from the bench, to the extent that it was.
6. Again it is submitted on the recusal application that these questions "suggested" certain things deprecatory of Tarrant. I reject this and do not see how a reasonable observer could so conclude. The questions were entirely open and non-leading. They sought explanation from the witnesses as to why she took one course and not another.
7. It is said on behalf of Tarrant that the questioning "suggested" that she had exaggerated to Roff the seriousness of the incident with Alois Rez and had manipulated Roff into thinking that the violence against her "not only on that night but on many previous occasions was much greater than it truly was". These "suggestions" are simply not to be found in the questions, to any reasonable reader of them. In this passage of evidence Tarrant was asked only about the events of the evening of 23 July 2013. The questions did not touch upon "many previous occasions".
8. Further, the questions elicited the explanation that the reason Tarrant had not phoned Roff again to tell him to stop the police attending her home was that he, Roff, "was telling me he (sic) Alois needed to be removed from the house if he had assaulted me". The questions afforded Tarrant the opportunity to explain that the eventual attendance of police, notwithstanding that all violence and threat of violence had ceased, was at the insistence of Roff himself not through any manipulation by her.
9. Tarrant's counsel further submits that the questioning at T 461.8 - .32 "demonstrates a very real lack of understanding of police protocols" regarding cancellation of domestic violence callouts. I do not consider that a reasonable observer could infer that bias against Tarrant with respect to her sentence should be apprehended from the asking of these questions simply because it may not have been possible for Tarrant or Roff to stop the police attending once the first call had been made. (That is what Tarrant's counsel asserts would be the effect of the "protocols").
10. It is submitted in support of the recusal application that the questions asked were irrelevant. I reject that. They went to the resolution of what otherwise was an anomaly in the narrative which the Crown saw fit to lead from this witness. Further, whether the questions were irrelevant to the issues in the Roff trial was a matter between the Crown and Roff. Neither objected to the questioning on the ground of relevance or sought to have the jury directed with respect to the answers obtained. It does not appear to me that a reasonable observer would infer judicial bias against Tarrant with respect to her sentence on the ground that irrelevant questions (if that is what they were) were asked of her in the trial of Roff.
11. Finally with respect to this passage of questioning from the bench it is submitted on behalf of Tarrant that it was undertaken "for the purposes of making an adverse finding against Tarrant in her own sentencing proceedings". This is another bald and unfounded assertion and I reject it.

The Roff trial - further questions on the events of 23 July 2013

1. Four further questions asked by me of Tarrant at T 462.10 - .23, again concerning the events of 23 July 2013, are relied upon as giving rise to an apprehension of bias in relation to her sentencing. These questions followed immediately upon the Crown having asked Tarrant what occurred when the police arrived at her home. She answered to the effect that she did not tell the police anything about Alois Rez pushing her down the hallway or otherwise acting violently because "I was too scared to tell them".
2. Given her evidence on the preceding two pages that the altercation between herself and Alois Rez had ended approximately two hours earlier and that in the meantime she had cleaned up, done the dishes and gone to sleep in the lounge room, the questions asked from the bench and now impugned were directed to resolving what appeared a contradiction in her account. Her answers did resolve that apparent contradiction. She explained that what she was "scared" of did not concern the altercation that had passed. Rather, she was afraid that more violence would be perpetrated if she reported the earlier events to the police.
3. With respect to this passage of questioning Tarrant's counsel now complains, firstly, that it "reveals a lack of understanding of the issues involving domestic violence generally, and the very real dangers that physically smaller and usually less strong women face". With great respect, the questions were directed to the specific facts of the trial in which they were asked not to "domestic violence generally". Being so directed, to a specific fact situation involving two particular individuals, Tarrant and Alois Rez, the questions were not capable of revealing "a lack of understanding" of a general social phenomenon such as "the very real dangers that physically smaller and usually less strong women face". I reject the contention that asking these questions in this fact specific context involved any formation or imparting of a view, accurate or inaccurate, regarding domestic violence in general.
4. Secondly it is submitted that the questioning "strongly suggests that Fagan J considered Tarrant's account of domestic violence at the hands of Alois Rez, not only on this night but in the nine years leading to that night, to be of little true consequence (contrary to the jury's verdict in the Tarrant trial)". I consider that it could not reasonably be apprehended by any informed observer that this questioning reflected any judgment or opinion about the significance of domestic violence of Alois Rez towards Tarrant over the nine preceding years. They were questions about a brief incident on a single evening in July 2013.
5. The questions undoubtedly constituted a testing of Tarrant's assertion that she remained "scared" notwithstanding that on her own description what had occurred that evening was a relatively brief and minor assault which had resolved some two hours before the police attendance. I do not consider that a reasonable observer could infer bias on the part of the trial judge towards the witness, relevantly to pending sentence proceedings against her, arising from the asking of questions which could reasonably appear to the trial judge appropriate to ask to explore an apparent improbability in evidence which she was giving adverse to the accused on trial.
6. It is further complained that one of the questions I asked rolled up three aspects of what Tarrant had described and sought her response to whether the altercation had been "a bit of a non event". It is now said that the answer "yes" is of doubtful meaning because it may constitute agreement with only one of the three elements of the question. The proposition that this question was "inadmissible for its triple propositions" has no bearing upon whether the questioner might reasonably be inferred to hold a bias against Tarrant with respect to her sentence. If uncertainty of the meaning of the answer arises from the form of a judge's question that again could not possibly lead to a reasonable apprehension of bias.

The Roff trial - questions regarding Centrelink payments

1. At T 547.15 - .43 counsel for Roff cross-examined Tarrant about her receipts from Centrelink. She said that she received approximately $1200 per fortnight and I then asked "In respect of how many children?" The answer was that she "had three in my name and Alois had one in his name". The cross-examiner then put to her that this was an arrangement "that you and Alois orchestrated between you … that you were claiming three of the four children", to which Tarrant assented. Counsel for Roff had available to him the transcript from the trial of Tarrant which recorded her evidence to the effect that Tarrant and Alois Rez had made such an arrangement between themselves for the purpose of maximising their receipts from Centrelink. Before arraignment and empanelling of the jury counsel for Roff had informed the Court that he had reviewed this transcript in the course of preparing for the trial of his client.
2. In the submissions to me that I should recuse myself it is acknowledged that questioning on this subject was likely to take place "given [Roff's counsel's] access to Tarrant's trial transcript". The passage complained of (at the transcript reference given in [47]) was immediately followed by questioning from Roff's counsel at T 547.44 to 548.21 referring to bank statements from the Commonwealth Bank and identifying three separate periodic payments that were recorded to the credit of those bank accounts "one by way of pension and the other two are not certain". Counsel went on to obtain from Tarrant an admission that she did not declare to Centrelink that she and Alois Rez were living together and that the claims were fraudulent, the payments being obtained in circumstances where she and Rez "consistently lied on declarations as to your living circumstances".
3. It is now submitted that my seeking clarification of what the Centrelink payment of $1200 per fortnight was for "prejudicially painted her in a poor light" and "gives rise to an apprehension of bias". On the contrary, that answer on its own merely clarified what the payment related to and neither suggested to the witness nor adduced from her anything concerning her honesty in dealings with Centrelink. The course of the cross-examination on credit by Roff's counsel which followed immediately thereafter shows that this was a fully prepared topic of questioning. The single question from the bench made no contribution to it whatsoever. I do not consider that any reasonable observer could apprehend bias on my part towards the witness in relation to her future sentence proceedings from the single question of factual clarification which I asked.

The Roff trial - questions regarding Tarrant's love of Roff

1. At T 578.34 - 580.7 Roff's counsel asked Tarrant a series of questions regarding the development of her romantic and sexual relationship with his client. She confirmed earlier evidence that when Roff had asked for her mobile phone number in December 2012 she had entered it in his phone and inserted the contact name "yes please". She agreed that in early 2013 she had sent Roff sexually explicit and provocative photographs of herself and that by March 2013 they had first had sexual intercourse in the back seat of a motor vehicle on the side of the road between Dubbo and Sydney. Then at T 580.9 - .17 there followed these questions: "[SUTHERLAND] Q. Did you love him? A. I don't know HIS HONOUR Q. Really Ms Tarrant? Really? You can't give a positive answer to that question having been reminded of those ways in which you communicated with him from the beginning of January 2013? A. At that point in time, yes."
2. This intervention would quite clearly have conveyed to the jury that I did not consider she was making a conscientious endeavour to answer counsel's question honestly and to the best of her ability. She responded "I don't know" to a question which, as it appeared to me, she would be well able to answer one way or the other having been reminded of the circumstances. Further background to the apparent evasiveness of this answer was that she had written the detailed and impassioned letter at Tab 8 in Ex C (see [22]), upon which she had answered questions at length in her evidence in chief, never suggesting that it misrepresented her true feelings.
3. Where a witness appears to a trial judge not to be endeavouring to answer a question truthfully to the best her ability it is open to the judge to press her to do so. It would alternatively be open to leave it to cross-examining counsel to insist upon a definitive answer. But where the witness concerned was the principal witness in the Crown case, in fairness to the accused I considered it appropriate to lend the authority of the Court to require her to give a more satisfactory answer.
4. Counsel for Tarrant on the recusal application asserts that my thus having intimated to the jury that I did not find her answer "I don't know" believable has given rise "to a strong suggestion of an apprehension of bias". I reject the proposition that by reason of a trial judge performing his duty in the conduct of a trial, in the respect referred to in the preceding paragraph, a reasonable observer properly informed could entertain a reasonable apprehension that that judge would hold a bias against the witness with respect to pending sentence proceedings concerning her.

The Roff trial - questions regarding Roff's threat to leave the district

1. At T 594.50 - .22 I asked questions of Tarrant arising out of her evidence concerning the terms of an ultimatum which Roff was said to have issued to her. Her account of this was that Roff had threatened "if Alois is not gone within three weeks" then he, Roff, would be "out of here". She attributed to the accused that he had made a threat to this effect and claimed that she had understood it as meaning he intended to leave the district in which he had lived all his life. Tarrant's counsel now complains that questioning at the above reference "made Tarrant appear to be either naive and silly, or a liar, yet it was the same evidence she had given in her own trial and which was obviously accepted by her jury."
2. I do not consider it to be inherent in the jury's verdict in the trial of Tarrant that they accepted her evidence, as given in that trial, that she believed Roff really intended to leave the Dubbo district if Alois Rez was not "gone" within three weeks from when this threat was conveyed. This incident, as described by Tarrant, was one of many events in the long narrative to which she testified at her own trial. It cannot be inferred from the jury's general verdict of guilty of manslaughter that they must have accepted as proved beyond reasonable doubt every single incident she described, in particular this one. But whether or not her evidence in that respect had been accepted by another tribunal on another occasion was no reason why an apparent improbability in this respect should not be tested with her by the judge presiding at Roff's trial. There was no objection from either the Crown or defence counsel to these questions and no direction to the jury was sought in relation to the answers. Further, Roff's counsel asked supplementary questions which resulted in Tarrant accepting that Roff may have intended only that he was threatening to leave his relationship with her.
3. A reasonably informed observer would consider that such questions from the judge in the trial of Roff constituted a permissible testing of an apparently improbable assertion made against the accused by a Crown witness. As such they reflected the performance of the judge's duty in the trial of Roff. If the questions had the effect of making the witness appear "either naive and silly, or a liar", that could not reasonably be taken to indicate bias against the witness in relation to her sentence proceedings.

The Roff trial - use of inappropriate language by defence counsel

1. At T 598.19 defence counsel asked a question of Tarrant in these terms: "Q. No, with the exception that he (scil, if he) buggered off and lived with Amanda Vine you would be quite happy about that, wouldn't you?"
2. I did not consider the use by counsel of the word "buggered" appropriate to the formality of the trial. I thought it was a departure by counsel from proper standards of court room behaviour. I also considered that it could reflect adversely on the Court if left without comment but that I should be cautious with respect to rebuking counsel in the presence of the witness and jury for such a choice of language. To do so might undermine counsel's authority in the middle of his cross-examination of the Crown's most important witness and might also diminish him in the eyes of the jury, to the detriment of the accused.
3. Very shortly afterwards at T 598.41 defence counsel asked this question: "Q. Let's not mince words, you were screwing another man?"
4. Within another page of transcript the morning tea adjournment was taken and at T 600 after the jury had been escorted out I directed counsel to keep his questions "to more formal language". He apologised for the lapse. I considered this a sufficient manner of dealing with a departure from court room standards. I was confident there would be no repetition. Had there been repetition I would have had to sacrifice the interests of the accused and his counsel by admonishing the latter in the presence of the jury in order to maintain the dignity of the proceedings. That did not occur.
5. It is now complained on the recusal application that this "use of colloquial language [i.e. 'buggered'] was disrespectful and counsel ought to have been reminded, immediately and in front of the jury, that the witness should not be spoken to in this manner". Further it is submitted that when the word "screwing" was used, the correction I made in the absence of the jury "should have been made immediately, and made with emphasis, in front of the jury. It was demeaning to Tarrant, in the context of criminal trial, for it to be put to her that she was 'screwing' a man other than her de facto husband". The point of these complaints about the manner in which I dealt with counsel's lapses into the vernacular is the submission that the "failure to ensure Tarrant was shown respect gives rise to an apprehension of bias on the part of Fagan J."
6. I reject that the trial judge's response to counsel's use of this language was controlled by the consideration of ensuring that Tarrant was shown respect. The issue was one of regulating the proceeding to ensure its formality and dignity and to maintain the respect of the jury and of the public generally for the manner in which the trial was conducted. As mentioned there were competing considerations affecting the severity with which and the circumstances in which it was appropriate to check counsel in his use of unacceptable language. I do not consider that any reasonable person could consider that the balance which I struck in dealing with this matter - allowing one use of inappropriate language to go unremarked and then correcting counsel, in the absence of the jury, on the "second strike" - in any way reflected any attitude on my part towards Tarrant or reflected any possibility of bias towards her with respect to the forthcoming sentence proceedings.

The Roff trial - question of Tarrant's "loyalty and adherence" to Alois Rez

1. At T 598.10 Roff's counsel asked Tarrant questions about Facebook messages she had seen indicating that Alois Rez intended to leave her and to live with another woman named Amanda Vine. Then at T 598.24 - .33 I asked questions as follows: Q. [Did] His apparent plan and intention to do that cause you to feel less loyalty and adherence to him in your relationship? A. Can you explain that in a way that I can properly understand? Thank you. Q. Did your discovery that he was exchanging messages with [Amanda] which indicated her very strong interest in him and his willingness to go along with it did that cause you to have less loyalty with Alois in respect to continuing your relationship with him? A. Yes."
2. As the evidence has been transcribed the words inserted in square brackets in the extract above are not included. Either those words were spoken and have been omitted in the transcription or, if they were not spoken, they reflect the sense of the questions asked. Tarrant's counsel complains on the recusal application that the first of these questions "was clumsy, as evidenced by Tarrant's response". I discount this criticism because it does not appear that ineptitude in a judge's questioning could give rise to any apprehension of bias, on any matter, against the witness to whom the poorly framed question was directed.
3. It is further submitted that: "the term 'adherence' suggests some old-fashioned form of servile connection, female to male. Against the background of the days of evidence Tarrant had already given in her own trial of the years of abuse she had suffered at the hands of Alois Rez, and the evidence she had already given in the Roff trial, the question of Tarrant's 'loyalty' to Alois Rez in this form was prejudicial. The whole of Tarrant's trial involved an examination of why she no longer felt 'loyal' to Alois Rez. The questioning carried the danger of suggesting that Tarrant's discovery of Alois Rez's interest in Amanda Vine and his thoughts of moving in with her were Tarrant's true motivation for wanting him dead."
4. This analysis is followed by the submission that the "questioning gives rise to an apprehension of bias". I do not consider that any reasonable observer would read into the word "adherence" any concept of an "old-fashioned form of servile connection". As it happens, the core of Tarrant's evidence about her mistreatment by Alois Rez was that he did maintain her in a servile relationship, belittling her and demanding that she perform menial household work whilst himself providing no domestic assistance and earning no income. It was she who asserted the servile relationship. It could not reasonably be taken from my questions that I suggested that was a fitting way for her to be treated.
5. The submission that questioning Tarrant about whether Alois Rez's disloyalty to her in conducting a liaison with Amanda Vine in turn caused her to feel less loyalty to him could not reasonably support the proposition that a fair minded observer would treat the question as manifesting a prejudice against Tarrant. The submission is that it was "prejudicial" "and that it "carried the danger of suggesting" a true motive for Tarrant to kill Alois Rez. This is an extravagant extrapolation from two simple questions concerning the effect upon Tarrant of discovering Alois Rez's infidelity.
6. Further, the language of this submission would be appropriate only if the questions had been asked in a trial of Tarrant herself. There was no question of prejudice or dangerous suggestion in relation to Tarrant's position before the jury because she was a Crown witness, not an accused. I do not accept that any reasonable observer properly informed could apprehend bias on my part against Tarrant with respect to the pending sentence proceedings on the basis of the strained reasoning in the submission quoted at [65].

The Roff trial - cross-examination of Tarrant on her physical relationship with Alois Rez

1. At T 580 - 581.7 counsel for Roff questioned Tarrant about her unsatisfying physical relationship with Alois Rez. She agreed with his proposition that Alois Rez was verbally demeaning with respect to her participation in sexual intercourse. He elicited evidence from her that Alois Rez described her as "starfish", meaning that she lay with her legs and arms apart and did not exhibit any physical responsiveness to love making.
2. No objection was taken to this line of questioning by the Crown. It was apparently conducted upon instructions. As trial judge I could not reasonably have been expected to have known, before the answers were given, what Tarrant would say in answer to counsel's questions. When the answers had been given the Crown did not seek to have the jury specially directed with respect to them and did not take objection to the line of questioning or suggest that it ought not be pursued further. In the event it was not.
3. Tarrant's counsel now submits on the recusal application: "This evidence was demeaning, disrespectful and belittling. It should not have been permitted. Although no objection was taken, Fagan J ought to have said clearly and firmly to Roff's counsel in front of the jury that questioning of this sort was not only irrelevant but it was impermissible and unacceptable. Tarrant was not represented in Roff's trial, so she did not have counsel acting in her interests to object [to] these questions being asked. For this questioning to go unremarked on by Fagan J gives rise to an apprehension of bias."
4. I did not consider that questions concerning Tarrant's unhappiness in her relationship with Alois Rez, generally, were irrelevant. A great deal of evidence was led from her by the Crown about numerous aspects of the relationship that made her unhappy and numerous ways in which Alois Rez demeaned and belittled her. The Crown lead such evidence as part of the narrative by which Tarrant explained what had brought her to agree with Roff that they would kill Alois Rez. This particular piece of demeaning was simply more in the same vein. I infer that the Crown did not object to it because, whatever may have been defence counsel's purpose, it was another item tending to confirm the powerfully motivating unhappiness of Tarrant's relationship with the deceased and thereby tending to support the likelihood of her having arranged to kill him.
5. The evidence being relevant and admissible, there was no basis upon which I could have intervened to say "clearly and firmly to Roff's counsel in front of the jury" that it was not. As for it being incumbent upon the trial judge to say that the evidence was "impermissible and unacceptable" I do not consider that there is any such independent ground upon which relevant evidence could be rejected and counsel chastised for attempting to adduce it. Any reasonable observer properly informed could not construe my refraining from having intervened with respect to this evidence as other than a performance of the duty of the presiding judge at the trial. Such an observer could not reasonably take from this an apprehension that my allowing the evidence to be received by the jury somehow involved a "demeaning, disrespectful and belittling" attitude on my own part towards Mr Tarrant or any other form of bias against her relevant to her sentence proceedings.

The Tarrant trial - Alois Rez's compliance with an AVO in 2008

1. It was a large part of Tarrant's defence case in her own trial that she sought to prove a number of instances of violence or threats of violence against her by Alois Rez. Her evidence to this effect was directed to establishing that the course of Alois Rez's treatment of her had induced an abnormality of mind which would support her defence under s 23A Crimes Act 1900 (NSW).
2. In support of the recusal application a general submission is made that questions I asked of Tarrant during her own trial suggested that she "had overstated Alois Rez's physical violence, verbal abuse and threats towards her". The first passage of my questions which is criticised in this respect is at T 701.19 - 703.1 and concerned why she had not considered leaving Alois Rez and going to a women's shelter, why she had not obtained an apprehended violence order ("AVO") against him and whether an AVO that had been taken out in December 2008 had been effective.
3. It is submitted on the present application that from my asking these questions I "did not appear to accept that Tarrant had good reason for thinking that Alois Rez would not abide the terms of any apprehended violence order imposed on him". A fair reading of the questions referred to in the preceding paragraph would show to any reasonable observer that all questions were in open form, non-leading except to the extent that they put back to her answers which she had already given in response to her own counsel. Bearing in mind that the witness being questioned was the accused in her own case constraints as referred to in the authorities at [10] and [11] were observed so as to give her an opportunity to explain her position and not to convey to the jury any doubt or judgment on the part of the trial judge about her veracity.
4. Notably, no objection was taken by Tarrant's counsel when these questions were asked. No request was made in the absence of the jury that I should desist from asking questions. No application was made that I should direct the jury in any particular way with respect to the questions already asked or the answers elicited. No specific direction about my having asked these questions was sought during the course of the summing up. Any reasonable and properly informed observer would identify from the absence of any such reaction from Tarrant's counsel at trial that the asking these questions did not signify any judgment being made of her or any bias against her in any respect whatsoever. Defence counsel immersed in the atmosphere of the trial was in the best position to sense any such implication from the questions. The absence of any reaction from defence counsel is the surest indication and would be to any reasonable observer that far from exhibiting bias against Tarrant these questions were within the proper scope of the trial judge's conduct of the proceedings.
5. The second passage of questioning impugned is at T 713.19 - 713.49. These questions were concerned with whether it was an option for Tarrant, if she left Alois Rez to live with her mother. Again the questions were non-leading and there was no objection or other reaction to them from defence counsel. It is now submitted on behalf of Tarrant that "Fagan J did not appear to accept that Tarrant thought Alois Rez's violence meant that it was very difficult for her to leave him and remain safe, if she took the children". Again, on a fair reading of the six questions asked in this passage no such appearance of a judgment in this respect having been made by the trial judge could reasonably be inferred.
6. It is submitted on behalf of Tarrant on the basis of these two passages of questioning that "Fagan J stepped into the arena of the trial, giving rise to an apprehension of bias". When the passages relied upon are read in context the submission that this involved the trial judge stepping "into the arena of the trial" is fanciful and I reject it. A comparison of the questioning which is now impugned with that in cases where it has been held that a trial judge has improperly "entered the arena" shows this submission to be out of touch with what appellate courts have to date regarded as the acceptable limits of judicial intervention in the questioning of witnesses. These two passages of judicial questioning were a far cry from the type of intervention which would characterise the trial judge as having been a combatant on the issues.
7. A third passage criticised in the submissions now made on behalf of Tarrant, at T 775.36 - 776.4, followed from evidence given by her that Alois Rez had said if she smoked cigarettes "he'd break your fingers". Immediately following this evidence I asked three non-leading questions as to whether Alois Rez had ever carried out this threat or one like it. These questions are the subject of submissions on behalf of Tarrant similar to those made in relation to the other two passages already discussed. I dismiss those submissions on the same basis as I have dismissed the equivalent submissions in the earlier paragraphs under this heading.

Proceedings after conviction in relation to Roff

1. The submissions on behalf of Tarrant seeking that I disqualify myself quote in full the transcript of proceedings against Roff after his conviction on 13 May 2016, from T 1.24 - 6.26. As that transcript will show I raised with counsel for the Crown and for Roff what I considered to be significant factual and procedural issues which would arise in the sentencing of Roff. First, I pointed out that Sarah Tarrant's evidence was that in discussion with Roff in May 2013 she made only a brief reference to her desire that Alois Rez be "gone" and that this had never been discussed again until four days before they killed Alois Rez when, on 25 July 2013, Raymond Roff provided her with sleeping tablets to administer to the deceased and formulated a specific plan for the killing. I expressed the view that this was an improbable sequence and pointed out that I had heard only Tarrant's version of how much discussion had preceded the homicide because Roff had denied any involvement at all.
2. It was important that counsel be forewarned of my scepticism about Tarrant's account in this respect because it could be an issue which Roff would wish to address in evidence now that he had been convicted. The extent of prior planning and the respective parts played by Roff and Tarrant, if that planning had been more extensive than as described by Tarrant, could be material to the degree of culpability of either or both offenders.
3. Secondly, I drew to counsel's attention my tentative impression that Tarrant's conduct on the night 23 July 2013 had been manipulative of Roff "to propel him" towards finalising a concrete plan for the elimination of Alois Rez. I expressly stated that this view was put "tentatively" and "argumentatively". It was important to forewarn the Crown and Roff of my initial reaction to the evidence because an ultimate conclusion upon it, again, could have a significant bearing upon the degree of culpability of each of the offenders respectively and relatively to each other. Discussion of this matter gave the Crown and Roff the opportunity to consider whether they would wish to seek to adduce any additional evidence relative to what I identified as a material factual question on sentence of Roff.
4. Relevant to the question of characterising the events of 23 July 2013 it was also important to indicate to counsel my understanding of her evidence given in her own trial regarding her past experience with an AVO against Alois Rez. Namely that he had complied with it. It was necessary that counsel were aware of my impression that actual physical violence against Tarrant, as recounted by her, had been moderate. These were aspects of the evidence which I considered would tend to the view that Tarrant had not really regarded herself as under physical threat on the night of 23 July 2013.
5. All of these matters were discussed with counsel expressly for Roff and for the Crown in tentative and argumentative terms, to put them on notice of factual issues which I considered important to the level of Roff's culpability and which counsel may wish to prepare themselves to address either in argument or evidence. I was at pains to state these issues in terms of matters which I would need to resolve factually on such admissible evidence as was available. No reasonable observer of this discussion could take from it that I had determined any of these issues in advance, in particular that I had determined any of them adversely to Tarrant. The whole point of the discussion was to flag issues which I had not resolved, to give an opportunity for more evidence on submissions to be brought to bear upon those issues. This is the antithesis of pre-judgment.
6. I also expressly pointed out to counsel that I faced a difficult task reconciling, on the one hand, the extensive evidence which the Crown had adduced in both trials of considerable premeditation, planning, rational action and the making of incidental moral judgments by Tarrant and, on the other hand, the jury's verdict which necessarily involved a finding that she suffered under an abnormality of mind at the time of the homicide. I made perfectly clear that I had not formulated any finding or made any pre-judgment on this issue but recognised it as a task lying ahead in the sentencing proceedings and an issue upon which "I must give her representatives an opportunity to speak". It was material to raise this question in the presence of counsel for Roff because his client would have an interest in it. The degree of responsibility which could be attributed to Tarrant, consistently with the jury's verdict, would impinge upon the corresponding degree of responsibility of Roff himself.
7. The outcome on the issue referred to in the preceding paragraph might affect the view taken of Roff's culpability and have a bearing upon his sentence. That was a significant reason why it appeared to me to be desirable that the sentence proceedings in respect of both offenders should be concurrent. That is why I drew the issue to the attention of the Crown and of Roff's counsel on 13 May 2016. Having done so I canvassed the possibility of a joint sentence hearing with both counsel immediately following my reference to the significant difficulty which I saw in determining the degree of Tarrant's responsibility.
8. The submissions on behalf of Tarrant on the application that I should recuse myself refer to a number of aspects of the discussion with counsel in the Roff proceedings, as summarised in the preceding paragraphs, as containing "central flaws". All of the arguments for disqualification based upon my exchange with counsel in the Roff post-conviction proceedings, at pars 48 - 51 on pages 21 - 24 of the Tarrant's written submissions, are matters to be raised and dealt with in the sentence proceedings themselves. Tarrant's counter arguments to propositions that I argumentatively canvassed with counsel in the Roff proceedings do not disclose anything about statements made by me on that occasion which could give any reasonable bystander an apprehension that I have prejudged anything in relation to the Tarrant sentence proceedings. The nature of statements made by me to counsel in the Roff proceedings on 13 May 2016 were explicitly the identification of issues to be resolved, the diametric opposite of pre-judgment upon any matter.
9. The discussion on 13 May 2016 with counsel for the Crown and for Roff was, in my experience, typical of a discussion between the bench and bar, in relation to a matter to be decided by the judge alone (in this case, penalty), when the evidence is complete or substantially so. By 13 May 2016 I had heard Tarrant's evidence in her own trial (308 pages over three days, equally split between evidence in chief and cross examination) and the same volume again in Roff's trial. In making tentative observations about aspects of her evidence and issues thereby thrown up I could hardly be said to be prejudging her. She had been extensively heard and tested in my presence.

Actual bias

1. In the penultimate paragraph of Tarrant's written submissions it is asserted that I have displayed actual bias in pointing out to counsel in the Roff case on 13 May 2016 the conflict between the jury's verdict and a great deal of the Crown's evidence, as identified at [86]. The suggestion of actual bias appears to be that I have predetermined that Tarrant suffered from no abnormality of mind sufficient to establish the partial defence under s 23A. It is not the case that I have made any such predetermination. On the contrary, I explicitly stated on 13 May 2016 at T 3.41 "she is to be sentenced on the basis that the jury found her of diminished responsibility".
2. At T 4.5 on 13 May 2016 I made clear that it lay ahead of me to determine how I would reconcile to this verdict the evidence tending to show her capacity for planning and her grasp of the difference between right and wrong: "I'm just pointing out there is one conundrum that I have in trying to sentence Sarah Tarrant. The jury has said she is of diminished responsibility and I see a raft of contemporaneous indicia of full mental capacity in every respect referred to in the section, that's why I want to hear from psychiatrists". These statements refute the pre-judgment alleged by counsel for Tarrant.

Conclusion

1. Accordingly the formal order of the court is: The application that Fagan J recuse himself from further hearing of sentence proceedings in respect of Sarah Tarrant following upon her conviction of manslaughter on 14 May 2016 is dismissed.

DISCLAIMER - Every effort has been made to comply with suppression orders or statutory provisions prohibiting publication that may apply to this judgment or decision. The onus remains on any person using material in the judgment or decision to ensure that the intended use of that material does not breach any such order or provision. Further enquiries may be directed to the Registry of the Court or Tribunal in which it was generated. Decision last updated: 28 June 2016