Folbigg v Attorney General of New South Wales

1. [1]

   THE COURT: The applicant, Kathleen Megan Folbigg, was convicted of offences with respect to the deaths of her four children. With respect to the first, Caleb, she was convicted of manslaughter. With respect to the next three children, Patrick, Sarah and Laura, she was convicted of the murder of each child. Further, with respect to Patrick, who, in the weeks before his death suffered an “apparent life-threatening event” (ALTE) the applicant was convicted of assault causing grievous bodily harm. She was duly sentenced to imprisonment and appealed against both her convictions and sentence. The appeal against the convictions was dismissed; the appeal against sentence was allowed and her sentence was reduced to imprisonment with a non-parole period of 25 years and an overall sentence of 30 years. [1] An application for special leave to appeal to the High Court was dismissed. [2]

2. [2]

   In 2018, the applicant petitioned the Governor seeking an inquiry into her convictions and her sentence. She said there was material which raised a reasonable doubt as to her guilt and as to the appropriateness of the sentence. The petition was made under s 76 of the Crimes (Appeal and Review) Act 2001 (NSW) (“Appeal and Review Act”). Having considered the petition, the Governor arranged for the Hon R O Blanch AM QC, a former judge of this Court and former Chief Judge of the District Court, to conduct an inquiry into the matters raised by the petition. (Adopting the terminology of the Appeal and Review Act, it is convenient to refer to him as “the judicial officer”.)

3. [3]

   On 22 July 2019, the judicial officer sent his Report to the Governor in accordance with s 82(1)(a) of the Appeal and Review Act. The Report stated that he had not formed an opinion that there was a reasonable doubt as to the guilt of the applicant, nor as to any matter that may have affected the nature or severity of the applicant’s sentence.

4. [4]

   On 21 October 2019 solicitors for the applicant filed a summons in the Common Law Division seeking review of the determination of the judicial officer in the supervisory jurisdiction of the Court under s 69 of the Supreme Court Act 1970 (NSW). At the time the summons was filed, the review jurisdiction conferred on the Court of Appeal depended on the decision-maker being, relevantly for present purposes, a judge. The judicial officer was a former judge. However, between the time the summons was filed and the hearing, the Supreme Court Act was amended so that the jurisdiction was conferred upon the Court of Appeal with respect to former judges. [3] The parties accepted that the summons should be transferred to this Court. It was listed before the Court of Appeal for hearing commencing on Monday, 15 February 2021.

5. [5]

   It is now some 18 years since her conviction and the applicant has served two-thirds of her non-parole period. It is, of course, desirable that the proceedings in this Court be disposed of with expedition. However, some explanation should be given of the significant lapse of time since her conviction. It is not the fault of any party or institution involved in the case. Rather, it is the result of a system of review which may be engaged at any time where doubts arise as to the soundness of a criminal conviction or sentence. The Governor’s Direction to the judicial officer identified the convictions and stated that the “doubt or question concerns evidence as to the incidence of reported deaths of three or more infants in the same family attributed to unidentified natural causes”. [4] That concern arose in large part as a result of medical and scientific developments post-dating the applicant’s trial.

6. [6]

   Although the applicant’s legal advisors had been able to obtain some material supporting the petition, further expert opinions were sought subsequent to the giving of the direction in August 2018 and reports were obtained by the Crown Solicitor’s Office (assisting the inquiry) from nine medical and scientific experts. The applicant obtained reports from 10 experts. Many of the experts gave evidence at the inquiry, as did the applicant. Directions as to the scope of the inquiry and procedural steps to be taken occurred on four occasions between 25 October 2018 and 11 February 2019. At the first directions hearing, counsel assisting submitted that the scope of the inquiry should be expanded to include:

7. [7]

   The judicial officer held hearings between 18 March 2019 and 1 May 2019 when the evidence concluded. Some six weeks after the conclusion of the evidence, the applicant sought to reopen the inquiry to tender further expert evidence. The final document provided to the inquiry was supplied on 28 June 2019. The basis of the reopening application will be dealt with below. It is a great credit to those responsible for conducting the inquiry that by 22 July 2019 the Report, the substance of which exceeded 500 pages, was able to be presented to the Governor.

Nature of judicial review proceeding

1. [8]

   The circumstances involving the deaths of four children and the imprisonment of their mother, in three cases having been convicted of their murders, were undoubtedly tragic and apt to arouse a high level of interest in the community. Roughly contemporaneously, there were two cases in the United Kingdom involving mothers convicted in respect of the deaths of children in which the convictions were set aside on the basis that expert evidence thought to have weighed heavily with the jury was either doubted or shown to be wrong. [6] There was an understandable concern that a similar injustice may have been done to the applicant.

2. [9]

   The thoroughness of the inquiry undertaken in 2018-2019 was attested to by the careful and comprehensive Report which resulted. The conclusion of the Report was that there had been no miscarriage of justice in this case. That conclusion cannot be overturned by this Court unless it was attended by legal error. The function of this Court is to ensure that the proper boundaries of the process were not exceeded and that the inquiry did not misconceive or otherwise fail to carry out the obligation of investigation and inquiry to the limits required by the law. The nature of the legal errors asserted by the applicant and the proper limits of judicial review will be considered further below. However, unless a relevant and material legal error could be identified, this Court is neither required nor permitted to intervene. It is no part of this Court’s function of review to form its own opinion as to whether a reasonable doubt attended the convictions of the applicant, and even if it were to form such a view it would not be entitled to give effect to it. The limited scope of this Court’s function is reflected in the fact that the review of the extensive inquiry was completed in two days of oral argument. Although there were both written submissions and oral submissions, it is apparent that the Court was not taken through each page of the Report, let alone the underlying evidential material. Rather, counsel for both the applicant and the Attorney General, correctly focused upon what could demonstrate legal error on the part of the judicial officer. Nevertheless, the Court has had the opportunity to read and consider the whole of the Report. For reasons explained below, the application must be dismissed.

Issues for consideration

1. [10]

   The grounds upon which this Court can undertake judicial review depend in part upon the nature of the function being exercised by the decision-maker. Thus, review of decisions made by a court may be more constrained than review of administrative decisions, because a judge exercising judicial power will have jurisdiction to decide legal issues necessary for the determination of a case. That is not necessarily true with respect to other decision-makers.

2. [11]

   For present purposes, it is not necessary to characterise the functions exercised by the judicial officer conducting the inquiry as either “judicial” or “administrative”: that dichotomy is probably inappropriate in this context. [7]

3. [12]

   The supervisory jurisdiction of this Court under s 69 of the Supreme Court Act is available to correct jurisdictional error, or error of law on the face of the record; it is not available to review exercises in fact-finding, or exercises of discretion vested by statute in the judicial officer, except to the extent that error in one of the two categories identified above is established. [8]

4. [13]

   There is no precise definition of what may constitute jurisdictional error. It depends upon the identification of the boundaries of the statutory power conferred on the decision-maker. [9]

5. [14]

   The second category of grounds, error of law on the face of the record, involves two components. The concept of legal error again may be separated into two parts. One part reflects the concept of jurisdictional error, namely identification of the boundaries of the power conferred. Such an error is reviewable regardless of whether or not it appears “on the face of the record”. However, especially with inferior courts, there is a second component. Whereas a court will be acting within jurisdiction in determining legal issues, if it mistakes the law it is applying, there will be an error of law within jurisdiction. Such an error, so long as it appears “on the face of the record”, will be amenable to review in the supervisory jurisdiction of this Court. Under the general law, the record was limited, in most cases, to the documents which engaged the jurisdiction of the tribunal and its formal order disposing of the matter before it. [10] At least in the case of tribunals and courts, s 69(4) of the Supreme Court Act extends the record to include reasons given by the tribunal for its order. It was common ground between the parties that the Report prepared by the judicial officer was part of the record and could be examined to determine if it demonstrated error of law. While inquiry before the judicial officer was not a proceeding in a court, it was accepted that he formed a “tribunal” for the purposes of s 69(4). However, it should also be assumed that the judicial officer had no authority to resolve legal questions erroneously. Thus, any material legal error is likely to be a jurisdictional error and is reviewable by this Court.

6. [15]

   There was a degree of ambivalence on the part of the applicant as to what precisely was the determination or decision to be reviewed. The summons seeking judicial review sought an order “quashing the report” and, in the alternative, a declaration that “the findings contained in the report” were legally flawed.

7. [16]

   Section 82 of the Appeal and Review Act requires that, on completion of an inquiry, the judicial officer must “cause a report on the results of the inquiry” to be sent to the Governor. No doubt, if the judicial officer failed to take the step of sending a report to the Governor, a mandatory injunction could be obtained directing the officer to take that step. Further, the language of s 82(1) may be understood as implying that a written document must be prepared in the form of a “report”. If no such document is prepared, again it may be possible to obtain a mandatory order that it be prepared. Neither of those issues arises in the present case. There is no suggestion that the document in fact delivered to the Governor was not a “report” satisfying the requirements of s 82(1).

8. [17]

   There are two possible ways of understanding the nature of the relief sought in the present proceeding. At a higher level of generality, it may be understood as a claim that the judicial officer has not completed an inquiry in accordance with law, so that the purported report does not satisfy the terms of s 82(1) as a report prepared on completion of a valid inquiry. Alternatively, and more precisely having regard to the consequences for the applicant, the relief sought is directed to the opinion expressed by the judicial officer that he was not satisfied that there was a reasonable doubt as to the guilt of the applicant. Indeed, it is that conclusion, and the intermediate findings and the reasoning of the judicial officer on which the opinion is based, that constitute the gravamen of the challenge. The latter, more particular, understanding of the nature of the claim is to be preferred, because it allows the review to focus on the scope of the available challenges.

9. [18]

   The fact that the case turns on a review of an opinion formed by the judicial officer has two consequences. First, it provides a basis for identifying the legal limits which constrain the formation of such an opinion; secondly, this Court has no power to form such an opinion for itself. The opinion-forming function is vested by statute in the judicial officer appointed to conduct the inquiry, and no one else.

10. [19]

    The legal constraints on powers to form opinions were helpfully identified by Latham CJ in The King v Connell; Ex parte Hetton Bellbird Collieries Ltd in the following terms: [11]

11. [20]

    These passages were relied on as forming the basis of the supervisory jurisdiction with respect to the review of powers conditioned upon the formation of an opinion by the decision-maker by Gummow J in Minister for Immigration and Multicultural Affairs v Eshetu. [12] Three propositions may be derived from this analysis.

12. [21]

    First, the decision-maker must identify the legal principle to be applied. If the relevant principle has been wrongly stated in a material respect, the court may intervene; if the principle has been correctly identified, as in the present case, it will be necessary to establish that the principle in fact applied was not that correctly identified. That may be apparent from the reasoning of the decision-maker, or may be the only proper inference from the decision itself in the light of the evidence available to the decision-maker. [13]

13. [22]

    Secondly, implicit in the proposition that the statute cannot have intended that the opinion be arbitrary, capricious or irrational, is the conclusion that it must be reached as a result of fair procedures. That approach was made explicit by the High Court in Re Refugee Review Tribunal; Ex parte Aala. [14]

14. [23]

    Thirdly, if it is possible, on the basis of the material available to the decision-maker to characterise the opinion in fact formed as arbitrary or capricious, or manifestly unreasonable, then the opinion may be held to be beyond power. Although Latham CJ did not use the language of unreasonableness, as opposed to irrationality, Gibbs J in Buck v Bavone [15] stated that the decision could be set aside if it appeared to the court to be “so unreasonable that no reasonable authority could properly have arrived at it.” One way of looking at that test is to infer, consistently with the second limb of the first proposition set out above, that the correct principles cannot have been applied.

15. [24]

    With those general principles in mind, it is convenient to identify the grounds alleged by the applicant as the basis for challenging the validity of the Report and the findings underpinning the judicial officer’s conclusion. So far as they could be distilled from the summons seeking judicial review, the grounds fell into two categories. First, it was alleged that the judicial officer failed to apply “the correct legal test”. This ground constituted the focus of the argument presented in this Court. The form the argument took will be identified below.

16. [25]

    The second set of issues was divided into the categories of “jurisdictional error” and denial of “procedural fairness”. The division was unnecessary: not only is procedural unfairness a category of jurisdictional error, but all of the errors alleged were in effect procedural errors. Further, there were various amendments made to the list of errors both prior to and in the course of the hearing. A convenient course is to list the procedural errors ultimately relied upon, putting to one side the form of the document provided to the Court:

17. [26]

    Even this restatement of the issues may be misleading in two respects. First, in identifying each as involving jurisdictional error, it was far from clear why, taken individually, any of the supposed errors would have demonstrated the invalidity of the final opinion, or the inquiry as a whole. [16] Secondly, each of the errors reads as the ground of an appeal from a trial judgment, where rulings have been made on the admissibility of evidence. This factor raises an aspect of the applicant’s case which will need to be addressed in some detail. Central to this issue, and the question of the application of the correct legal test, was a fundamental proposition that the inquiry had to be conducted as an inquiry into the guilt of the applicant by applying the principles governing a criminal trial, but without any constraint on the evidential material.

18. [27]

    There was a similar related issue which arose in the course of the hearing. The applicant relied in her summons (and her list of issues) on an allegation that the complaints as to the conduct of the inquiry demonstrated “predetermination or apprehended bias” on the part of the judicial officer. This ground was problematic in more than one respect and was abandoned in the course of the hearing. However, the basis of the ground was that the judicial officer had started with the assumption of guilt, because that was the finding of the jury, which was upheld in the Court of Criminal Appeal, and then asked whether he was satisfied by further material that there was a reasonable doubt as to the applicant’s guilt. Although not properly characterised as prejudgment or apprehended bias (or indeed actual bias) the underlying complaint remained alive as part of the legal question as to the correct approach to be applied by the judicial officer in the conduct of the inquiry.

Justiciability

1. [28]

   The Attorney took no issue with the jurisdiction of this Court to review the determination of the judicial officer made at the completion of the inquiry. Nevertheless the Court should be satisfied that it has the relevant jurisdiction. Whilst this matter was pending in the Court decisions of the Queensland Court of Appeal and the Full Court of the Federal Court of Australia reached a different conclusion with respect to an equivalent Queensland provision. It is, therefore appropriate to state why the Court is content to accept the Attorney’s submission that it has jurisdiction.

2. [29]

   The first of the two cases, being a judgment handed down on 12 August 2020 by the Queensland Court of Appeal, was Holzinger v Attorney-General of Queensland. [17] The application in Holzinger was made pursuant to s 672A of the Criminal Code (Qld), which provides:

3. [30]

   Mr Holzinger’s petition sought a pardon or, alternatively, a referral of his petition to the Court of Appeal. There were two decisions in response to the petition, namely (i) a determination by the Governor not to pardon him, and (ii) a decision by the Attorney-General not to refer his petition to the Court of Appeal. It was the latter decision which was the subject of the application for judicial review. [18]

4. [31]

   There are three matters to be addressed in respect of this case. First, although the application was made only under the Judicial Review Act 1991 (Qld), nothing seems to have turned on the question whether a decision not to refer a case was a decision of an administrative character made under an enactment. [19]

5. [32]

   Secondly, the conclusion that the decision of the Attorney-General was not amenable to review was supported on two bases. The first was that a power conferred on the Attorney-General “to commence litigation” was not amenable to judicial review. [20] That reasoning concluded:

6. [33]

   The second basis of rejection was that the statutory power to refer a case to the Court of Criminal Appeal was an incident of the pardoning power, the exercise of which was itself not amenable to judicial review. [21]

7. [34]

   The second case was Attorney-General (Cth) v Ogawa, [22] judgment being delivered by the Full Court of the Federal Court on 28 October 2020. Dr Ogawa, having been convicted in Queensland of offences under the Criminal Code Act 1995 (Cth), invoked the procedures available under s 672A of the Criminal Code (Qld). It was accepted that the powers to be exercised were those of the Governor General and the Commonwealth Attorney General. (It appears to have been assumed that s 68 of the Judiciary Act 1903 (Cth) operated to confer on Commonwealth officers powers to deal with a petition arising under the Queensland Criminal Code. [23] ) Given that the decision in Ogawa on appeal post-dated the Queensland Court of Appeal decision in Holzinger, the Full Court considered itself bound to follow Holzinger with respect to the operation of the Queensland statute. [24]

8. [35]

   There are three aspects of these cases which remove any obligation for this Court to apply them so as to deny jurisdiction to review the determination of the judicial officer. First, the statutory scheme under the Appeal and Review Act is similar to, but not identical with, that found in s 672A of the Criminal Code (Qld). The Queensland section refers to a “petition for the exercise of the pardoning power”, as do equivalent provisions elsewhere in Australia). Rather, s 76 of the Appeal and Review Act, dealing with petitions to the Governor, is in the following terms:

9. [36]

   Secondly, Holzinger and Ogawa were directed to a decision not to refer a case to the Court (there being no statutory power to require an inquiry); the closest analogy under the Appeal and Review Act would be a decision not to direct an inquiry by a judicial officer under s 77(1)(a). That step having been taken in the present case, in accordance with the applicant’s request, the exercise of that power is not in issue.

10. [37]

    Thirdly, reliance on authorities which deny the availability of the supervisory jurisdiction with respect to decisions to initiate prosecutions against a person have no obvious application to a decision by a government officer to exercise a power to permit an inquiry at the behest of the convicted person. The statute creates in the Governor and the Minister functions of a gateway to the operation of the statutory scheme. That is, they can allow a convicted person an inquiry to review his or her conviction, a function which is the antithesis of prosecuting the person for the purpose of obtaining a conviction.

11. [38]

    Those reasons are sufficient to preclude reliance on the cases referred to as authority to deny jurisdiction in this Court to review an inquiry which has in fact been held.

12. [39]

    There is in fact authority which would support the amenability of the inquiry to judicial review. Section 140 of the Sentencing Act 1994 (WA) contained a provision in similar terms to s 672A of the Queensland Criminal Code. Mallard v The Queen [25] involved an appeal from a case referred to the Court of Appeal under that provision. The Court of Appeal had dismissed the petition. It might be thought curious that a petitioner could enforce in the High Court his legal right to a proper consideration of his petition by the Court of Appeal, but could not have reviewed the refusal of the Attorney General to refer the matter to the Court of Appeal. In any event, Mallard tends to support the view that, having obtained a direction that a judicial officer hold an inquiry, the applicant was entitled to invoke the supervisory jurisdiction of this Court to ensure that the inquiry was properly conducted in accordance with the statute.

13. [40]

    The Court may be comfortably satisfied it has jurisdiction to consider the present application.

Recurrent infant deaths in one family

1. [41]

   Before identifying, by way of background, the circumstances of the children’s deaths, it is convenient to note a submission which never descended into detail, except as a matter of history, but which has been raised at various stages of the judicial hearings, the inquiry and the present application for review. Thus, in the written submissions in this Court it was said that the petition which gave rise to the Governor’s direction and the inquiry claimed that a doubt arose as to the applicant’s convictions because of “[t]he application of the now discredited Meadow’s Law”. [26] Further, the submissions stated that it was “clear from a close examination of the Inquiry report that the method of reasoning deployed in the consideration of the medical and scientific evidence closely followed the judgment of Justice Sully [in the Court of Criminal Appeal rejecting the appeal against convictions in 2005].” [27] After setting out a passage from the conclusions in that judgment, the submission continued, stating that the judicial officer “failed to grapple … [with] [t]he debunking of ‘Meadows Law’ ….” [28]

2. [42]

   Chapter 4 of the Report contained a lengthy discussion under the heading “Recurrence”, introduced by the following statements:

3. [43]

   Sir Roy Meadow was a British paediatrician; his “law” was explained in the following passage in Ch 4:

4. [44]

   In the history of the trial proceedings, this form of reasoning appeared to be adopted by the prosecution in resisting an application for separate trials with respect to each death. It relied in part on the evidence of Dr Janice Ophoven, an American paediatrician, who gave evidence in the following terms: [35]

5. [45]

   Although this Court was not taken to evidence as to the basis of such a statistical calculation, it presumably operated on an assumption that each of the deaths was an independent event. Such reasoning would not only have been wrong but incoherent. The support for an unnatural cause relied upon the element of common parentage; the possibility of a genetic cause arose from precisely the same common element.

6. [46]

   Counsel for the applicant before the Court of Criminal Appeal in 2003 (the appeal from the rejection of the separate trials application) was alert to the problem. He expressly objected to reliance upon the views of medical experts “on matters of probability and statistics”, noting that similar reasoning had resulted in the English Court of Criminal Appeal overturning a conviction in the matter of Sally Clark. The judgment in that case was apparently not then available, and he invited the Court to withhold judgment in the matter before it until the report was available. [36]

7. [47]

   The invitation to delay was not accepted, Hodgson JA (with the agreement of Sully and Buddin JJ) considered that the primary judge (Wood CJ at CL) “was careful to base his decision only on so much of the medical evidence as was not in any way affected by statistical considerations in such a way as to be liable to be rendered inadmissible.” [37]

8. [48]

   At the trial, which occurred between 1 April 2003 and 21 May 2003, the prosecution did not adduce the opinions of Dr Ophoven in evidence.

9. [49]

   The Report set out the evidence given at the trial relevant to this issue. [38] The Report then set out both the investigations conducted by research scientists for the inquiry, and evidence given by clinicians at the inquiry. [39] The Report recorded in some detail a history of the recurrence of such deaths in the literature, both prior to 2003 and subsequent to that date. As noted above, the Report observed that the reasoning of Professor Meadow “has now been widely discredited” and referred to the case of R v Cannings. [40] That decision of the English Court of Criminal Appeal followed the decision in R v Clark.

10. [50]

    Cannings had been decided prior to the applicant’s appeal against her convictions and was discussed in the judgment of the Court of Criminal Appeal in 2005. [41]

11. [51]

    The inquiry dealt further with later cases, both in England and Victoria. Submissions by counsel assisting in the following terms were then set out:

12. [52]

    The final conclusions of the inquiry were expressed as follows:

13. [53]

    In the course of oral submissions in this Court, senior counsel for the applicant accepted that the judicial officer had described the reasoning as “discredited” and did not adopt it: no error was identified, as appears from the following exchange between the bench and senior counsel (Mr Morris SC). After taking the Court to a letter written by Professor Meadow in January 1989, stating that “it is important to recognise that a small proportion of children labelled ‘SIDS’ are killed by their parents, usually their mother”, counsel continued: [48]

14. [54]

    It is apparent that the judicial officer did not accept or adopt Prof Meadow’s discredited approach. At trial the prosecutor adopted a more nuanced approach, and the Court of Criminal Appeal in 2003 and 2005 also eschewed the discredited approach.

Background

1. [55]

   Because the challenge mounted by the applicant drew heavily on the manner in which the judicial officer had dealt with the medical and scientific evidence in relation to the deaths of each child, it is necessary to provide some factual background with respect to those deaths. The background may be succinctly stated as it is set out in significant detail, not only in the Report (which is publicly available) but also in decisions of the Supreme Court with respect to an application for separate trials in respect of each death, [49] the appeal against that decision, [50] the first appeal from the verdict [51] and the reopened appeal judgment. [52]

(a) first child – Caleb

1. [56]

   Caleb was born on 1 February 1989. He died 19 days later on 20 February 1989. The cause of death was, in effect, identified as unknown by the entry on the death certificate “SIDS”, namely sudden infant death syndrome. As explained by Duncan and Byard the term:

2. [57]

   This terminology indicates that death was sudden in the sense that there were no known material pre-existing illnesses or natural events causally connected to the death, and that there was no evidence of unnatural events, such as smothering. Because it is “a diagnosis of exclusion”, any medically accepted explanation takes the death out of the category. The incidence of such diagnoses reduced following a public “Back to Sleep” campaign, and should continue to reduce incrementally as the rate of affirmative diagnoses improves. [54]

3. [58]

   The Report recorded the following medical history: [55]

4. [59]

   Two days later he was found dead in his cot by the applicant. She later told police that she had fed him at 1:00am and put him back in his cot. A diary entry at about 2:00am identified that he was “finally asleep!!”. She checked him at about 2:50am and found him cold and lifeless. When the first ambulance officers arrived at 2:59am, he was unconscious, not breathing and pulseless. The officer noted that Caleb was warm to touch and pale around the mouth and lips.

5. [60]

   A forensic pathologist examined Caleb within 12 hours of his death and found no external signs of injury and no evidence of abnormality. A toxicology report was negative. An examination of the lungs noted:

6. [61]

   The evidence at trial was, in broad terms, that the condition of “floppy larynx” or laryngomalacia, was not a recognised cause of death, many of the medical experts not having heard of a child who had died from such a condition. The treating paediatrician, Dr Springthorpe, gave evidence that stridor, noisy breathing, had nothing to do with Caleb’s death.

7. [62]

   There was further consideration of the medical evidence at the inquiry, but without any definitive result. It appears to have been largely accepted that accidental asphyxiation was excluded but that deliberate asphyxiation was not. Whether any natural cause of death rose to the level of a reasonable possibility was a matter for evaluative judgment. A finding that there was no possible natural cause raised by the expert evidence was reasonably open.

(b) second child – Patrick

1. [63]

   Patrick was born at full term with normal weight on 3 June 1990. Because of Caleb’s death, a “sleep study” was conducted at one and half weeks: the results were normal. On 18 October 1990 (at age four months and 15 days) Patrick suffered an apparent life-threatening event. The applicant found Patrick in his cot at around 4:30am “because she heard Patrick gasping.” She said he was blue around the lips, lifeless and floppy and making minimal respiratory effort. Soon after, Patrick gave “a high pitched cry.”

2. [64]

   The ambulance officers arrived at 4:41am. They described him as in “respiratory distress”, with very laboured breathing. He was administered oxygen and taken to hospital. The applicant told the officers that “Patrick had had a cold for two days but was normally happy and dynamic.” The report recorded Patrick’s skin temperature as hot. The treating doctor at the hospital noted that he was lethargic, cyanosed and responsive only to painful stimuli on arrival, but after the administration of oxygen for 15 minutes he became more alert and remained pink. The doctor concluded that his condition “was not likely to be due to a respiratory problem.” The Report further noted: [56]

3. [65]

   Some 36 hours later, his temperature went up and he was vomiting and commenced having fits that evening. He remained in hospital until 29 October, when he was discharged, and was then readmitted on 4 November with a seizure. An electroencephalogram (EEG) on 5 November revealed abnormalities.

4. [66]

   Dr Dezordi said that the damage to Patrick’s brain “definitely” could have been caused by asphyxiation. [57] The Report’s summary of the evidence at trial continued:

5. [67]

   Patrick died on 13 February 1991. At 10:00am that morning, the applicant telephoned Mr Folbigg at work and screamed down the phone, “it’s happened again” and “I need you. Come home.” He drove home quickly, raced into Patrick’s room, scooped up Patrick from the cot and took him to the lounge and commenced CPR. He described him as “floppy, warm, with blue lips.” Ambulance officers arrived at the house at 10:10am. The final autopsy report included the following passages:

6. [68]

   The body of medical evidence was that the initial life-threatening event was a “catastrophic asphyxiating event” which could have been caused by deliberate smothering or by some unknown cause. At the inquiry, Professor Ryan identified various potentially causative diagnoses, including conditions possibly associated with epilepsy and fluctuating neurologic symptoms not excluded by previous testing.

7. [69]

   The judicial officer reviewed a considerable body of medical evidence, resulting in the following conclusions:

8. [70]

   The judicial officer, after expressing caution in accepting Prof Ryan’s evidence, particularly in relation to the possibility of genetic disorders which he considered had been the subject of extensive inquiry and rejection, [58] concluded:

9. [71]

   With respect to his death, the judicial officer made the following finding:

(c) third child – Sarah

1. [72]

   Sarah was born at full term on 14 October 1992. As with Patrick, Sarah was the subject of a sleep study carried out at three weeks of age. Dr Cooper, who had also undertaken the study with Patrick, considered the results to be normal for her age. A general practitioner who saw Sarah four times believed her to be a normal healthy infant. She had suffered from a flu or cold virus for which medication had been prescribed on 18 August 1993, with the final dosage taken on 26 or 27 August, three days before her death.

2. [73]

   At the time of death she was in a single bed in her parents’ bedroom. She was reported to have been snoring when her parents went to bed at 9:30 or 10:00pm. The applicant told police that she had got up at 1:30am, and, not hearing Sarah breathing, had turned on the light and saw she had turned blue and noted a discharge from the nose.

3. [74]

   At autopsy, two puncture abrasions were noted on her bottom lip, together with a scratch on her upper arm. Her uvula (the piece of tissue at the back of the throat which prevents food and liquid entering the nose) was reddened, a condition consistent with a mild infection or with snoring. The summary in the Report continued:

4. [75]

   Professor Hilton, who conducted the autopsy, listed the cause of death as SIDS.

5. [76]

   In chapter 5 of the Report, the judicial officer made the following findings:

6. [77]

   This, as the applicant noted, was not the final form of the intermediate finding of fact as to the possibility of a natural cause for Sarah’s death. That finding was to be found in the Addendum to the Report, which took account of information supplied some weeks after the hearings were completed on 1 May 2019. However, that matter concerned genetic information and a particular genic variation common to the applicant and the two girls, Sarah and Laura.

7. [78]

   In order to understand the significance of particular findings it is necessary to have regard to the structure of the Report. Chapter 4 was headed “Sudden Infant Death Syndrome and Recurrence”. The medical evidence in Ch 5 dealt with what was, in effect, a series of forensic pathology reports, followed by evidence as to smothering. That in turn was followed by Ch 6, dealing with the role of infection in SIDS, and Ch 7 dealing with genetics. Findings were made at the end of each chapter in relation to the children. An alternative structure might have addressed changes in different elements of the medicine before providing a global assessment of how those changes might affect each of the children. The purpose of identifying genetic variations which might have pathological consequences was to identify a natural causal explanation of one or more of the deaths. The structure which was adopted, on the other hand, gave rise to a cumulative process of assessment and reassessment at particular stages of the analysis. However, no complaint was made as to the structure of the Report (except with respect to the Addendum) and it is by no means clear that any plausible challenge could have been raised on this basis. It is sufficient to note that none of the findings outlined above was treated as final, or even as an intermediate, finding of fact, although as can be seen from the last set of findings regarding Sarah, reference to the genetic testing was taken into account before Ch 7, where the genetics evidence was set out.

(d) fourth child – Laura

1. [79]

   Laura was born at full term on 7 August 1997. A sleep study undertaken 12 days post-partum showed mild central apnoea, which was identified as “totally normal” by February 1998. The sleep study also identified no bradycardia in response to the earlier apnoea nor, by February 1998 were there any cardiac problems.

2. [80]

   Laura died around midday on 1 March 1999, aged 18 months. She was found by the applicant, who called the ambulance service. Ambulance officers arrived at 12:14pm and found that Laura was not breathing and had no pulse. There was no blood, vomit or foreign object in her mouth. Her skin was warm to the touch and cyanosis was present around the lips and face. A record of her immediate history indicated that she had had a cold in the last 24 hours, had been administered Demazin two days earlier on 27 February and had been last fed at 7am on 1 March. She had been put down to sleep and was on her back when she was found. There was nothing covering her head. Given her age, she did not fit within the usual criteria for SIDS and, together with her age, the testing and monitoring led to her being considered to be at extremely low risk for SIDS. [59]

3. [81]

   Dr Cala, who performed the autopsy found there was “an inflammatory infiltrate in the heart consistent with myocarditis”. [60] However, having watched a video of her playing taken on 28 February 1999 (the day before her death), Dr Cala considered she appeared to be in good health and considered that the myocarditis “played no role whatsoever in her death”. [61] There were a number of further medical expert reports prepared with respect to Laura, based on the more comprehensive material that was available relating to her death, as compared with the deaths of the other children. This material was considered at length by the judicial officer who was satisfied that, based on the medical evidence received at the inquiry, “the myocarditis found in Laura’s heart at autopsy could have been incidental to her death, or it could have been fatal.” [62] He also noted that deaths of children from myocarditis are rare and there was about one death each year in New South Wales. Sudden and unexpected deaths are fewer again and more frequent in young babies than in the age range of one to four years. He also accepted that “[n]o forensic pathologist has excluded the possibility of an unnatural cause of Laura’s death”, [63] concluding that it was “also reasonably possible that her death was caused by an event leading to obstruction of her airways, which includes deliberate smothering.” [64]

(a) Immunology

1. [82]

   Chapter 6 of the Report was headed “Immunology”. It addressed evidence from microbiologists suggesting that infection may have played a role in the deaths of one or more of the children, and evidence concerning research suggesting that “slight infection” was a “risk factor” associated with SIDS. The question for the immunologists was whether infection may have played a role in the deaths of any or all of the children. They also considered whether, in cases of SIDS, a genetic factor, such as a pathogenic genetic variant in genes associated with immunological responses, may have increased the risk of an otherwise apparently minor infection triggering a fatal outcome.

2. [83]

   The judicial officer considered that “[t]he evidence of the microbiologists goes no further than to speculate that infection may have played a part in some of the Folbigg children’s deaths.” [65]

(b) Genetics

1. [84]

   Of more importance to the outcome of the inquiry was the discussion of “Genetics evidence” in Ch 7. In so far as the evidence established a particular natural cause as a reasonable possibility in relation to the death of each child, taken separately the likelihood of such natural causes, each in itself a rare event, in four consecutive children in one family was vanishingly small. The conclusion would, of course, be quite different were there a real possibility of a common cause. The common cause might, in theory, be environmental or genetic. In practical terms the question was whether there was evidence to support a common genetic link.

2. [85]

   There was no doubt that there had been significant advances in the field of genetics since the trial in 2003. There was available to the inquiry material taken from each of the children, from which DNA could be extracted. In December 2018 the applicant provided a sample for genetic testing. The inquiry engaged a multi-disciplinary panel of experts which undertook further genetic sequencing and analysis. Two experts based in Canberra were retained for the applicant. It is convenient to deal with the evidence addressed in Chapters 6 and 7 when dealing with the specific particulars of review directed to their evidence.

3. [86]

   However, before dealing with the more detailed issues, some further general observations are in order. First, although the discovery of a plausible cause of death in any of the children would in theory reduce the degree of improbability of four deaths occurring naturally in one family, unless the causal hypothesis for a natural cause in one case was strong, it would have limited effect. Similarly, different indications in each of the four cases, lacking strong causal links, would have limited effect. However, the identification of a common genetic factor providing a possible explanation as to why more than one death occurred in the one family, would be, subject to the other evidence, an important step in demonstrating a reasonable doubt as to the applicant having caused those deaths.

4. [87]

   Secondly, despite extensive investigations, no common genetic element has been identified as present in all four children. What was identified, after the trial, was a variant in the CALM2 gene associated with certain cardiac conditions, the genetic variant being common to the girls (Laura and Sarah) and the applicant. As will be noted below, the causal significance of this was contestable.

5. [88]

   Thirdly, the identification of a relevant genetic factor associated with disease or illness and hence death will depend upon whether there is evidence in the clinical condition of the child or person (the phenotype) consistent with such known conditions. The clinical information relating to all four children included (i) the absence of pregnancy related complications; (ii) the absence of congenital malformations; (iii) newborn screening reported as normal; (iv) development normal for age for each child prior to death or the ALTE at four months for Patrick; (v) none had a surgical procedure, or was admitted to hospital with a significant medical problem (excluding Patrick’s ALTE); (vi) none was on continuous medication or had more than eight documented respiratory infections a year and (vii) all had autopsy examinations with no medical cause of death determined. [66] These circumstances largely dictated the basis of the trial in 2003. The hypothesised natural causes of death were individual for each of the children, namely laryngomalacia (floppy larynx) (Caleb), infection (Patrick), uvulitis / laryngospasm (Sarah) and myocarditis (Laura).

6. [89]

   Fourthly, as to genetics, in 2013 it was reported in the medical literature that genetic perturbations in CALM1, 2 and 3 which encoded for a particular protein, calmodulin, were associated with early occurrence of life-threatening cardiac arrhythmias in infancy and early childhood, particularly in long QT syndrome (LQTS [67] ) phenotypes, and less frequently in catecholaminergic polymorphic ventricular tachycardia (CPVT), idiopathic ventricular fibrillation (IVF) and sudden unexplained deaths (SUD) in young individuals’ phenotypes. [68] Thereafter, the authors of the report set up an international calmodulinopathy registry which collected demographic, clinical and genetic information in relation to patients suffering from LLQTS and CPVT. In April 2019 a paper was published relating to 74 subjects from 51 different families. [69] The significance of the 2019 paper (which became available after the evidence to the inquiry was completed but before the Report was released) lay in its identification of infant deaths associated with a variant in CALM3 which encoded the same protein as the CALM2 variant found in the applicant, Sarah and Laura . Prior to publication of the 2019 paper, the experts who gave evidence to the inquiry were generally not persuaded that this variant was likely to be pathogenic, although Professor Vinuesa stated that she “would not feel comfortable with excluding its potential for pathogenicity”. [70] The later publication led to further medical evidence and an addendum to the Report. The significance of the material was that it provided a common genetic link associated with pathogenicity in the two female children. However, the males were not affected. There therefore remained at least three separate causes, as there was no common link between the deaths of the boys.

(c) The diaries

1. [90]

   It is important to identify the contents of Ch 8, entitled “The Non-Medical Evidence including Ms Folbigg’s Diaries”. That evidence covered (i) the content of the diaries; (ii) attempts to conceal the existence of the diaries; (iii) her statements to the police in relation to the content of the diaries; (iv) her evidence to the inquiry as to the content of the diaries, and (v) evidence before the inquiry as to her attempts to dispose of the diaries. The evidence of Mr Craig Folbigg together with some brief evidence from other members of the family and other lay witnesses, including neighbours and friends, was also addressed in this chapter. It is sufficient for present purposes to note their evidence in relation to the diaries. At trial the prosecution asserted that various entries in the diaries contained “virtual admissions” by the applicant of her guilt for the deaths of three children with admissions that she was at risk of acting in the same way in relation to Laura.

2. [91]

   There was no dispute as to the authenticity of the diaries, nor that the entries were made by the applicant. The sole question was how particular entries should be understood. If they were to be understood as referring to the applicant having killed her children, directly, and were not delusional, they constituted powerful evidence of guilt which could not readily be overcome by medical evidence as to a possibility that each child died of natural causes. The judicial officer accepted, as a matter of fact, after hearing evidence from the applicant who had not given evidence at her trial, that the diaries should be read as including “virtual admissions” as to causing the deaths. This finding significantly contributed to the conclusion that he was left with no reasonable doubt as to her guilt of the five crimes for which she was convicted.

3. [92]

   An important element of the applicant’s case in this Court was that a different reading of the diary entries was available and, indeed, was the only reading reasonably available if one approached them with the knowledge that there was indeed a reasonable possibility that each death resulted from natural causes. Counsel submitted that, it being a circumstantial case, the judicial officer “had to exclude any available alternative interpretation that may be consistent with innocence in order to use the diary contents as a virtual admission.” [71] This was because the inferences drawn from the diary entries were “more akin to a link in the chain” than to a strand in a cable. [72]

4. [93]

   Before addressing this submission, it is necessary to identify briefly the subject matter under consideration, namely particular entries in the diary.

5. [94]

   Several significant diary entries commence in June 1996, some three years after Sarah’s death and a little over a year before Laura’s birth.

6. [95]

   A significant aspect of the applicant’s case was that these entries could all be read as regretting things she did not do, rather than as a record of things she did do. It was submitted that the judicial officer failed to give proper consideration to the omission/commission distinction.

(a) application of correct legal test

1. [96]

   The first four issues identified for determination by the Court involved the application of the correct legal test of “reasonable doubt” in relation to the evidence. It may be accepted that a material error in the application of the law would constitute a reviewable error which could lead to the quashing of the final conclusion. The problem is to identify the nature of the correct legal test: the concept of reasonable doubt is not readily open to further definition.

2. [97]

   There is no challenge to the articulation by the judicial officer of the proper approach to his function in carrying out an inquiry under Pt 7 of the Appeal and Review Act. In formulating the approach adopted, the judicial officer relied for each of the propositions contained in the following passage on statements in reports of earlier inquiries, and particularly on those of Wood J in an inquiry conducted in 1985. [73] The relevant passages in the reasoning of the judicial officer were as follows: [74]

3. [98]

   What was required of the judicial officer was essentially an exercise in fact-finding. That was the exercise undertaken. The purpose and subject matter of the Report were reflected in the consideration of extensive evidence as to whether there was a real possibility that one or more of the deaths had resulted from natural causes, so as to provide a factual basis for a finding of reasonable doubt as to the applicant’s guilt with respect to one or any of the charges. The application of the standard of “reasonable doubt”, as found in s 82(2)(a) of the Appeal and Review Act, was to inform the process of reasoning, and to form the basis of the final conclusions. Reading the Report as a whole permits no inference that the judicial officer deviated from that function.

4. [99]

   In oral argument, senior counsel for the applicant focused upon two propositions. The first submission was that “the medical evidence established a potential natural cause of death for each of the children that needed to be excluded if there was to be a finding that there was no reasonable doubt about the guilt of Ms Folbigg for the purposes of the Act.” [75] There followed an acknowledgment that the judicial officer had excluded those reasonable possibilities, on the basis of the diary entries, based on their implicit admissions. However, counsel submitted that those entries were themselves open to a reasonable innocent explanation, so that they could not have justified dismissal of the reasonable possibility of natural causes established by the medical evidence. That in turn led to a consideration as to whether each was in the nature of a link in a chain, or rather comprised strands in a cable, to adopt the metaphor used in Wigmore on Evidence, as explained by Dawson J in Shepherd v The Queen. [76]

5. [100]

   The second, related, submission was that the judicial officer needed to consider the entries in the diary from a neutral position as to whether the children died naturally or were murdered. The submission continued: [77]

6. [101]

   It is true that the metaphor of a chain or a cable identifies an important distinction in the way circumstantial evidence may operate. However, the metaphor has limitations. Similarly, there would, of course, be an element of circularity in adopting a reading of the diary entries based upon an assumption that no natural cause had been established for any of the deaths. However, neither of these critiques of the reasoning can be accepted.

7. [102]

   Both ways in which the applicant’s case was put (links in a chain and circularity) involved an assumption that there was a staged process in assessing pieces of evidence. That assumption was inapt in the circumstances. The process was nuanced and involved weighing of all the evidence. There were many factors involved. For example, there was no single standard of “reasonable possibility” in considering possible natural causes. The rarity of certain occurrences could not be ignored. The likelihood of a particular cause arising in a specific case required an evaluative judgment, based on an assessment of differing medical opinions.

8. [103]

   Similarly, an understanding of the meaning of the diary entries was not to be undertaken on a day by day basis, but having regard to knowledge of the applicant and her personal history and circumstances, and the explanations of them advanced by the applicant in her oral evidence to the inquiry.

9. [104]

   No error of law was demonstrated by the manner in which the judicial officer assessed the evidence in order to reach the negative conclusion that he was not satisfied there was reasonable doubt as to the guilt of the applicant in respect of any of the convictions. The concept of “reasonable doubt” was not a legal standard: if it called for a particular form of reasoning, it was a holistic approach to the evidence. The staged approach based on independent assumptions proposed by the applicant was inappropriate: the judicial officer’s conclusion depended upon a synthesis of competing and complicated factors.

10. [105]

    Of course, that was far removed from the exercise required of the medical experts: each properly gave evidence from the standpoint of his or her particular expertise. Each properly expressed views as to possible causes of death. It was not within their role to reach a conclusion as to whether reasonable doubt attended the convictions. Indeed, for them to have done so would have undermined their professional objectivity and might have diminished the force of their opinions. While the judicial officer was conscious of the disparate functions, he did not identify lack of objectivity affecting the expert witnesses at the inquiry.

(b) failure to accept listening device transcripts

1. [106]

   This, and a number of other grounds, were formulated as if the inquiry had been a trial in which admissible evidence had been wrongly rejected, giving rise to a miscarriage of justice. These grounds paid insufficient attention to the facts that (i) this was not a trial but an inquiry conducted under Pt 7 of the Appeal and Review Act, (ii) the rules of evidence were not applicable and (iii) any challenge to the procedure adopted by the judicial officer needed to be formulated in appropriate terms. An appropriate ground therefore needed to identify a relevant issue raised by the terms of the inquiry and then to identify evidence relating to an essential element of the inquiry, the rejection of which would demonstrate a constructive failure on the part of the judicial officer to carry out his proper function according to law.

2. [107]

   The particular error was said to be a failure to accept the tender of a transcript of a recording by a listening device placed in the lounge room of the Folbiggs’ home in July 1999. The listening device had been placed there pursuant to a warrant obtained by police in the course of investigating the deaths of the children. The conversation between the applicant and her husband concerned a possible response to the stress being caused by the police investigation. Mr Folbigg began an explanation of how he could have killed the children. The applicant interrupted at one point to say “ridiculous”. Mr Folbigg responded:

3. [108]

   The fact that the judicial officer accepted only a redacted version of this transcript as part of the evidence at the inquiry reflected a conclusion that the redacted part was irrelevant. It could only have been relevant to support an inference that the children were in fact murdered, not by the applicant but by her husband. Reading the whole transcript it is not possible to conclude that Mr Folbigg was purporting to claim actual responsibility for the deaths. Nor was it proposed by any party to the inquiry that such a finding should be made. In considering suggestions of procedural unfairness, the judicial officer recorded that there was no application by the applicant, “or indeed anyone else, to have Mr Folbigg called as a witness in the inquiry. In any event, this was beyond the scope of the inquiry and there was no application to extend the scope in this way.” [78] In final conclusions the judicial officer stated: [79]

4. [109]

   This ground of challenge must be rejected: no error of law, nor indeed any other kind of error, has been established in relation to the redaction of the listening device transcripts.

(c) failing to accept the tender of the unredacted Goldwater report

1. [110]

   The report in question was prepared by Professor Paul Goldwater and dated 29 March 2019. The applicant’s written submissions referred in three brief paragraphs to the fact that redactions had been made by the Crown Solicitor’s Office, “in spite of the [applicant’s] solicitor raising concerns with the Crown Solicitor about the proposed redactions.” The submissions identified five such redactions, each of which contained criticism of a report by an expert who had given evidence, those reports being annexures to a report of Professor Blackwell of 5 March 2019 which the solicitors for the applicant had supplied to Professor Goldwater. The applicant’s written submissions stated that the redacted portions “went to the credibility and accuracy of professional expertise of evidence given at trial and postulated natural causes of death for three of the children.” [80] The rejection of the evidence was said to amount to “procedural unfairness, and a practical injustice to the plaintiff.”

2. [111]

   The history of the redactions was explained by the Attorney in his written submissions in this Court, none of which was disputed by the applicant in reply or orally. The Goldwater report had been served on 29 March 2019, after the scheduled timetable and after experts to whom Prof Goldwater was responding had already given evidence.

3. [112]

   Counsel for the Attorney submitted that there could be no procedural unfairness because there was an opportunity for the applicant to have raised the issue of redaction before the judicial officer and seek to tender the additional material. In her reply, the applicant submitted that “the tender was complete”, and “[t]he direction was deficient to enable a revisitation as suggested.” [81] The submissions further stated that the redactions were made “unilaterally, without reasons and without invitation for argument before the judicial officer.” [82]

4. [113]

   The only available inference is that the applicant’s solicitors chose not to pursue the issue before the judicial officer. There may have been good reason why that was so: the report contained what might have been considered tendentious comments on the conduct of the trial and the expertise of witnesses, and the suggestion that parts of the evidence were “irrational”. These opinions would not obviously have assisted the judicial officer. In any event, in the absence of evidence that no application to tender the additional material would have been entertained, there was no breach of procedural fairness.

(d) failure to accept the tender of the unredacted Clancy report

1. [114]

   The submissions with respect to Professor Clancy’s report were substantially similar to those with respect to Prof Goldwater’s report. The response must be the same. There was no tender of the unredacted reports of Prof Clancy: therefore they were not rejected.

2. [115]

   Two reports provided by Prof Clancy dated 13 March and 17 March 2019 were not the subject of redaction. Most of the redacted portions of Professor Clancy’s third report related to evidence given by Professor Horne. It is apparent from a paragraph which was not redacted that Prof Horne had already given evidence. [83] No doubt any critique of matters raised by Prof Horne could have been the basis of submissions (the applicant’s lawyers having the full report by Prof Clancy), if they were already the subject of evidence. If not, reliance on Prof Clancy’s observations would have necessitated the recalling of Prof Horne. The judicial officer might have declined to allow a process by which an expert was entitled to comment on the evidence of others who had already been called, thereby possibly necessitating their recall. On the other hand he might not. The submissions before this Court do not reveal any application having been made either to tender an unredacted version of Prof Clancy’s third report, and, if necessary, to have Prof Horne recalled. There is no evidence to suggest lack of opportunity to take such a step. Prof Clancy’s third report was dated 27 March: the inquiry did not terminate until 1 May 2019. As will be seen, the applicant’s lawyers appear not to have been reticent about seeking to tender further material after 1 May 2019.

3. [116]

   There is no basis for a complaint of procedural unfairness on this ground.

(e) redacting reports by Professors Goldwater and Clancy prior to accepting their tender

1. [117]

   This ground is in effect a complaint about the process for tendering evidence adopted by counsel assisting. It provided no basis for asserting procedural unfairness to the applicant beyond those which have already been addressed.

(f) failure to reopen the inquiry after receipt of a late material from Professors Schwartz and Vinuesa

1. [118]

   On 21 June 2019, weeks after the closing of evidence and submissions, Professor Vinuesa sent the inquiry a further statement, attaching an advance copy of a 2019 paper by Lia Crotti et al. [84] Also attached to Prof Vinuesa’s statement was a letter to her from one of the authors of the Crotti paper, Professor Peter Schwartz.

2. [119]

   As has been noted, two main reports on genetic analysis were provided to the inquiry, one by those known as the “Sydney team” and one by the “Canberra team”, the latter being Professors Vinuesa and Cook. The latter were nominated by the applicant’s lawyers. The subject-matter of the late material concerned a specific issue relevant to the girls, the background to which should be explained.

3. [120]

   Where there is more than one case of SIDS in a particular family, one may suspect genetic factors at work. The research has tended to focus on genes relating to metabolic disorders, cardiac disorders, serotonin production (which regulates breathing among other functions), inflammatory responses to infection, development of the autonomic nervous system and seizures. As the judicial officer recorded, [85] neither the Sydney nor the Canberra team found variants in genes which were assessed as pathogenic or likely pathogenic in all four children so as to explain their sudden deaths. The genetic evidence in the present inquiry focused particularly on variants of genes associated with cardiac disorders. One target of investigation was the gene known as CALM2. The Report noted in Ch 7: [86]

4. [121]

   Prof Schwartz noted in his 20 June 2019 letter that the report he had seen (being the report prepared by the Canberra team) “only linked the CALM2 variant to a long QT syndrome phenotype, ignoring the possibility that phenotype could be CPVT. He noted that without an exercise stress test of Ms Folbigg, a diagnosis of CPVT was still ‘fully on the table’.” Although Professors Vinuesa and Cook had apparently not considered expression of CALM2 in CPVT, it is clear from the material set out above that the inquiry had evidence before it which did consider that possibility. The evidence was also that the applicant had undergone an exercise stress test: Prof Skinner had formed the view that she did not have CPVT.

5. [122]

   The CALM2 variant detected in Sarah, Laura and the applicant encoded for the amino acid at the site glycine 114, known as Gly114. The significance of the Crotti paper, according to Prof Schwartz, was that the study group included “a family with an asymptomatic mother carrying a variant in the CALM3 gene affecting the same amino acid as was identified in the applicant, Sarah and Laura, namely Gly114, in which one child died at age five from a cardiac arrest while playing, and another died suddenly at age four”..” [88]

6. [123]

   On receipt of Prof Vinuesa’s further statement, counsel assisting provided the Sydney team with a copy and asked for a response. A short supplementary report was provided on 5 July 2019. The supplementary report agreed that the three CALM genes encoded an identical protein and that there was an increasing likelihood that the variant found in Sarah and Laura might be relevant to their deaths. [89] It concluded that if the clinical information were not taken into account, the variant would now be classified as “likely pathogenic”, meaning that there was a greater than 90% chance of it being disease-causing. On the basis of the clinical information, however, they concluded that “the fact Ms Folbigg is alive (at age 52), has never had a cardiac arrest, and produced exercise test results within normal limits are strongly against a hypothesis of concealed CPVT.” [90] The authors considered that the CALM2 variant remained “of uncertain significance, due to the conflict between the clinical and genetic evidence.” [91]

7. [124]

   Because the applicant contended that the fresh evidence should have been sufficient to persuade the judicial officer to reopen the whole inquiry, it is desirable to identify aspects of the response provided by Professor Kirk, Dr Buckley and Professor Skinner which supported the view that a high level of uncertainty remained as to whether the discovery of the CALM2 variant in the applicant and her daughters was of any causative significance.

8. [125]

   First, they addressed the nature of the variants:

9. [126]

   Secondly, they noted the significance of the variants in the Folbigg family being inherited:

10. [127]

    Thirdly, they referred to the significance of the children being found deceased during a sleep period:

11. [128]

    Whilst the authors accepted that the variant could be pathogenic and could be either the sole cause of death for Sarah and Laura or a contributing cause, they also noted the possibility that the variant was benign, stating:

12. [129]

    The Report noted that counsel assisting had been provided with an opportunity to file further written submissions, which they did on 8 July 2019. Counsel assisting contended that the correct finding remained that “there is no reasonable possibility that the death of any of the Folbigg children or Patrick’s ALTE was caused by a recognised genetic variant.” [92]

13. [130]

    The applicant’s representatives were given an opportunity to file submissions in reply and filed both a further written submission and a letter said to have been drafted by Professor Vinuesa and Dr Arsov with assistance from Professors Schwartz, Cook and Overgaard. The letter stated, apparently for the first time, that the registry of calmodulinopathies referred to in Crotti et al recorded “five cases of sudden death/cardiac arrest while asleep caused by CALM1/2/3 mutation.” However, the judicial officer noted that the letter did not suggest that any of those cases involved an infant. [93] After setting out more of the submissions for both parties, the judicial officer dealt at some length with the further information in making findings with respect to the CALM2 variant. He accepted “the opinion in the letter from Professor Vinuesa that the CALM2 variant identified in Ms Folbigg and Sarah and Laura ‘is likely to cause a cardiac condition amidst the CPVT/LQTS/IVF spectrum’.” [94] He further observed:

14. [131]

    His dispositive reasoning on this material was as follows:

15. [132]

    Returning to the complaints raised by the applicant based on this material, it is not correct to say that the inquiry was not “reopened” after receipt of the material from Prof Vinuesa on 28 June 2019. It was reopened; the material was all considered, as was further material supplied by her and further submissions obtained from both parties. The judicial officer considered whether it was appropriate to have a further oral hearing and determined that it was not.

16. [133]

    The written submissions for the applicant stated that procedural fairness required that she have the opportunity to test the further opinion provided by the Sydney team, which “may have affected their earlier evidence.” The submission further stated that the judicial officer “failed to reconsider his findings and conclusions that formed the main body of his report”. [95]

17. [134]

    None of these propositions should be accepted. Procedural fairness did not require that there be a hearing at which witnesses would be recalled: the judicial officer was entitled to call a halt to the inquiry, without being obliged to reopen the inquiry because the applicant had located fresh scientific information. Relevantly, the judicial officer considered the possibility of a further hearing: he noted that the “conflict in views and the debate it generates will no doubt continue, however, it is not necessary for it to be resolved in order for me to decide if the Inquiry should be reopened.” [96] He was also satisfied that no further genetic testing would materially affect the available information. [97]

18. [135]

    There was no submission (nor could there have been) that the procedural approach adopted by the judicial officer was manifestly unreasonable in the legal sense.

19. [136]

    The further suggestion that the judicial officer should have reconsidered the whole of the genetic evidence and the non-medical evidence in the light of the new material is without foundation. The judicial officer, fully cognizant of the findings which had already been made (and which were adverted to in the addendum), identified the significance of the further material, varied the finding to which it was specifically relevant, but expressed satisfaction that the result did not affect the outcome of the weighing of all the material which had already been undertaken. It is not possible to discern legal error in that approach.

(g) failure to consider the submission of Prof Clancy

1. [137]

   On 27 June 2019, the solicitor for the applicant provided to Prof Clancy a copy of the material provided to the inquiry by Prof Vinuesa on 21 June. On 28 June 2019 Prof Clancy sent to the inquiry an unsigned document described as a “Submission to the Inquiry”. In it he stated:

2. [138]

   Somewhat stridently, the applicant’s written submissions stated:

3. [139]

   It was also stated that the failure to address this material “demonstrated unwillingness by the Judicial Officer to be impartial and review matters upon which he had already had [sic] a concluded view.” [99] A separate ground relating to bias was withdrawn in the course of the hearing: it is assumed that this last unfounded submission must also have been withdrawn, as it properly should have been.

4. [140]

   Whilst the inquiry was not a judicial process, the judicial officer was entitled, if not obliged, to establish procedures for dealing with the significant volume of expert evidence which needed to be addressed. Prof Clancy was not a party to the inquiry: he had no right to expect that an unsolicited “submission” would be taken into account. That the submission went from him directly to the inquiry may be inferred from the content of the email which he sent to the solicitor for the applicant on the same day, namely Friday, 28 June 2019 which simply stated:

5. [141]

   If the applicant sought to draw some support from the document, her legal representatives should have sought leave to rely upon it. In truth, it is likely that Prof Clancy thought that his expertise (though his acknowledged speciality was immunology, not genetics) was intended to give weight to the opinions expressed in the submission. It was therefore further evidence and one can understand why there was no attempt to tender it. On 12 July 2019 the applicant’s solicitor filed a last submission headed “Addendum” which referred to the submission by Prof Clancy of 27 June 2019 in a footnote.

6. [142]

   It may be noted that Dr Cala and Professor Hilton made submissions on 14 June and 18 June respectively, which appear not to have been directly addressed by the judicial officer. They, however, had been given leave to appear before the inquiry.

7. [143]

   If the applicant complains that there was procedural unfairness to her in the judicial officer not expressly referring to the reasons of Prof Clancy, relied on by her in a footnote to a late submission, without quotation and without further explanation, then the submission must be rejected.

(h) failing to consider evidence as to the good character of the plaintiff contained in (i) statements by lay witnesses and (ii) the report of Dr Diamond

1. [144]

   A particular of challenge to the Report which was not pressed alleged that there had been a failure to consider the evidence of Dr Diamond in relation to the review of the applicant’s convictions. That particular was properly withdrawn: it could not have been drafted by a person who had read the Report. There were at least a dozen paragraphs and numerous footnotes referring expressly to the psychiatric report of Dr Diamond. Further, Ch 2 of the Report, entitled “Overview of the Folbigg Family”, contained much material favourable to the applicant which was sourced to Dr Diamond’s report, which is referred to in some 50 footnotes.

2. [145]

   Dr Diamond’s report was also referred to in considering how the diaries should be understood, in Ch 8 dealing with “non-medical evidence”. That, perhaps, should not be understood as a form of good character evidence, but one would not normally expect a professional report of a forensic psychiatrist to be considered as character evidence in relation to the subject of the report.

3. [146]

   The applicant’s written submissions commenced with the statement that “[g]ood character can be a consideration in criminal proceedings.” That is true, if an accused person seeks to adduce character evidence. The submission continued: [100]

4. [147]

   The applicant’s submissions did not identify the good character evidence, nor did they refer to any submissions made to the inquiry as to good character evidence. The Attorney inferred that the intended reference was to three affidavits which had been provided prior to the hearing, but which were not tendered. No response was made to that suggestion in the applicant’s submissions in reply.

5. [148]

   It should be noted, however, that the Report contained summaries of the evidence given by members of the family, neighbours and friends, much of which was either neutral or favourable to the applicant. What was sought to be made of this material by the applicant, if anything, was not identified in this Court.

6. [149]

   The precise nature of the error relied upon was not specified in the written submissions: however, the ground need not be considered further because the factual propositions underlying it are not established.

(i) failing to address the applicant’s submissions as to the interpretation of her diaries

1. [150]

   The content of the submissions which were said not to have been addressed by the judicial officer were not identified in this Court. However, the written submissions said to relate to this ground appeared to have two propositions underlying them. One appeared in the following statement: [101]

2. [151]

   This submission relied upon the proposition discussed above and rejected, that the judicial officer was bound to form a view that there was a reasonable alternative inference available from the diaries which was exculpatory, or neutral, rather than inculpatory, because the premise was an assumption of innocence.

3. [152]

   The second proposition appeared in the following passage: [102]

4. [153]

   The factual premise for this complaint was that relevant diary entries, and other evidence as to her state of mind, were excluded from the evidence taken at the inquiry. That premise was not made good. If the proposition was that counsel assisting did not tender all the material which the applicant thought relevant, then it was a matter for the applicant to take up at the hearing and proffer further material. However, as noted by the Attorney in his submissions, a 1992 diary identified by senior counsel for the applicant as relevant, which counsel assisting did not consider relevant, was in fact included in the bundle. [103] The applicant’s submissions in reply did not cavil with that proposition.

5. [154]

   In Ch 8 of the Report, express reference was made to Ms Folbigg’s sworn evidence about the interpretation of the diaries; [104] the submissions were summarised, [105] a complaint that she had been denied the opportunity to provide “context” evidence was specifically addressed, [106] and rejected, with explanation, as “without merit”. [107] The judicial officer’s construction of the diaries was set out in some detail. [108] That discussion expressly addressed proposed interpretations, suggestions made in her evidence and the various innocent explanations provided for specific phraseology. The factual proposition underlying this ground is without any foundation. The ground need not be entertained further.

(j) constraining the admission of evidence of context in relation to the diaries

1. [155]

   As noted in relation to the previous matter, this was an issue raised before the inquiry and expressly dealt with in the Report under the heading “Findings regarding procedural fairness”. [109] The judicial officer found:

2. [156]

   The written submissions for the applicant did not address these passages in the Report. In the absence of any submission that the statements contained in these paragraphs are inaccurate, the ground need not be entertained further.

(k) method of redacting documentary evidence

1. [157]

   Finally, the applicant maintained her reliance upon two allegations of procedural unfairness, namely the unilateral redaction of documentary evidence without consultation with parties given leave to appear, and “redacting documentary evidence that was relevant to the statutory function of the [judicial officer to inquire].”

2. [158]

   In substance, the question of redaction has been addressed in dealing with the matters identified under grounds (c), (d) and (e) above. Any concern that the process of redaction adopted by counsel assisting was procedurally unfair was a matter which should have been raised with the judicial officer in the course of the inquiry. It is the denial of an opportunity to raise an issue which is the gravamen of any complaint of procedural unfairness.

3. [159]

   In written submissions filed on behalf of the applicant it was said that the redactions “precluded the Plaintiff making submissions about those matters. This was an egregious denial of procedural fairness.” The colourful language does not overcome the absence of substance in the submission. It was then said that relevant documents had not been disclosed to the applicant, a claim unrelated to the ground. In fact it was not submitted that any document was withheld from the applicant: the redactions were all apparent on the face of documents which were tendered. In some circumstances it appears that the applicant took issue with the redactions, sometimes unsuccessfully; in other circumstances she did not. The ground is not an objection to rulings of the judicial officer, nor to the failure of counsel assisting to disclose documents which should have been disclosed. The fact that counsel assisting considered it appropriate to redact passages from particular documents involved no element of procedural unfairness. The applicant was not deprived of any opportunity to raise the completeness of the evidence with the judicial officer and obtain rulings as she or her legal representatives thought appropriate. The ground must be rejected.

Conclusions

1. [160]

   Had any of the plethora of specific complaints had been upheld, but others not, there would have been difficult questions as to the consequence of such rulings. Issues as to materiality of specific errors with respect to the operative opinion would have arisen. As the High Court held in Minister for Immigration and Border Protection v SZMTA [110] “[a] breach is material to a decision only if compliance could realistically have resulted in a different decision.” [111] None of this was addressed in submissions. As it is, each of the grounds pressed by the applicant and identified in the further amended list of issues handed up at the hearing, must be rejected.

2. [161]

   This was not a case in which the judicial officer’s conclusion was at odds with the scientific evidence. The scientific evidence raised a theoretical possibility that there were innocent explanations for the deaths of the two girls. However, the judicial officer was required to consider evidence that, although the CALM2 abnormality in Ms Folbigg and the two girls involved a change in an amino acid in the vicinity of Gly114, their circumstances departed from the reported cases of deaths associated with CALM abnormalities. In particular, (i) the change in the affected amino acid was not the same, (ii) the change in the Folbiggs’ genome was hereditary and not de novo, (iii) the girls died at younger ages, (iv) the girls apparently died suddenly when asleep and not during exertion, (v) there was an absence of prior symptomatology, and (vi) Ms Folbigg did not have LQTS or CPVT, being cardiac manifestations commonly associated with it. The girls’ deaths were thus “outliers” when compared with those reported in the literature. Further, the boys’ genomes provided no common cause. When these matters were weighed with the inculpatory inferences derived from Ms Folbigg’s diary entries and her evidence in seeking to present innocent explanations of them, there was an ample basis, consistent with the scientific evidence, for the judicial officer to conclude that there was no reasonable doubt as to Ms Folbigg’s guilt.

3. [162]

   The Court makes the following orders:

   1. (1)

      Dismiss the further amended summons seeking judicial review of the Report of the judicial officer.

   2. (2)

      Order that the applicant pay the Attorney’s costs of the proceedings in this Court.