Mudaliar v State [2008] FJSC 25; CAV0001.2007 (17 October 2008)

THE SUPREME COURT OF THE FIJI ISLANDS
AT SUVA

CRIMINAL APPEAL NO CAV0001 OF 2007
(Fiji Court of Appeal No AAU0032 of 2006)

BETWEEN:

SACHIDA NAND MUDALIAR
Petitioner

AND:

THE STATE
Respondent

Coram: The Hon Justice Keith Mason, Judge of the Supreme Court
The Hon Justice Kenneth Handley, Judge of the Supreme Court
The Hon Justice Mark Weinberg, Judge of the Supreme Court

Hearing: Tuesday, 14th October 2008, Suva

Counsel: Mr J. Haigh QC and Mr R. Naidu for the Petitioner
Mr A.G. Elliott for the Respondent

Date of Judgment: Friday, 17th October 2008, Suva

JUDGMENT OF THE COURT

1. The petitioner, a medical practitioner who, for many years, practised as an obstetrician and gynaecologist, was convicted in the High Court in Suva, in 2006, on one count of manslaughter. He was sentenced to three years imprisonment, which sentence he has now served. The charges against him arose out an allegation that he carried out an unlawful termination of pregnancy on Poonam Pritika Kumar on 21 March 2003. Poonam died shortly thereafter.
   
2. The offence of manslaughter is defined in s 198 of the Penal Code. That section relevantly provides:
   

"Any person who by an unlawful act or omission causes the death of another person is guilty of the felony termed manslaughter ..."

3. The count on which the petitioner was convicted was drafted in accordance with the suggested forms of information in the schedule to the Criminal Procedure Code. The particulars of offence stated simply that the petitioner:
   

"between the 21st day of March 2003 and the 22nd day of March 2003, at Suva in the Central Division, unlawfully caused the death of Poonam Pritika Kumar".

The proceeding before this Court

4. The Court has before it two separate petitions, each seeking special leave to appeal from decisions of the Court of Appeal. The first of those decisions was given on 24 November 2006 and will be referred to hereafter as "the fresh evidence judgment". The second was given on 23 March;2007 and will be refe referred to as "the substantive judgment".
   
5. By the fresh evidence judgment the Court of Appeal refused the petitioner leave to adduce fresh evidence at the hearing of his appeal against conviction. By the substantive judgment it dismissed the petitioner’s appeal against conviction.
   

Circumstances surrounding the offence

6. Poonam was aged twenty-two at the time of her death. She was a university student. In March 2003 she found that she was pregnant. The father, Abhikesh Kumar, was a fellow student.
   
7. The prosecution case was that on Wednesday, 19 March 2003 Poonam consulted a general practitioner, Dr Hazratwala who after ning hing her concluded that she was in the twentieth week of pregnancy. According to Dr HazratwPoonam wanted a term termination. Dr Hazratwala referrr to the petitioner.
   
8. Later that say, P and Abhikesh atth attended at the petitioner’s surgery. According to Abhikesh, PoonaPoonam made it clear to the petitioner that she did not want to have the baby. The petitioner prescribed some medication, and told her to return on the following Friday, 21 March 2003 when, according to Abhikesh, the petitioner agreed that he would carry out an abortion.
   
9. According to Abhikesh, both he and Poonam went to the petitioner’s surgery at about 2 pm on 21 March 2003. Abhikash was told that the procedure would take some hours. He left, and returned at about 9 pm. His nce was that at that that stage Poonam’s clothing was blood stained, and she was uncous. It should be noted that that two receptionists called by the State gave an entirely different account. They said that Poonam was, at that stage, conscious and able, with their assistance, to walk to the recovery room. Their version of events accorded closely with that of the petitioner.
   
10. Abhikesh said that the petitioner asked him to stay with Poonam for about two hours, and then take her home. However, Abhikesh was unwilling to do so because Poonam resided in a hostel which closed at 11 pm. The petitioner tuggestegested that Abhikesh should collect Poonam at 7 am the fong morning. It wast was common ground that the petitioner subsequently left Poonam on her o the recovery room of his surgery, locked the premises and and went home.
    
11. Abhikesh returned the following morning at 8 am to find the surgery still locked. He telephoned the petitioner and was told to wait for the receptionist to arrive. When she came, she let Abhikesh in. Together they found Poonam lying on the bed, dead.
    
12. Subsequently the petitioner arrived and confirmed that Poonam had indeed died. According to Abhikesh, the petitioner asked him to tell the police that Poonam had had a miscarriage. That was, in fact, what Abhikesh initially told the police when they arrived.
    
13. The medical evidence called on behalf of the prosecution was to the effect that Poonam had died from medical shock caused by a haemorrhage resulting from a severe tear in the uterine wall. The medical witnesses observed that that injury had also led to an air embolism. They said that in their opinion the embolism had resulted from a termination which had been badly performed.
    
14. The petitioner’s own case was presented essentially in the form of a lengthy and detailed unsworn statement. He called no other witnesses. He said that when he first examined Poonam on 19 March 2003, he concluded that she was about twelve weeks pregnant, but that she was in the process of miscarrying at that time. He gave her medication to prevent infection and told her to return on Friday. By then, according to the petitioner, she had fully miscarried. As a result he carried out a procedure, commonly known as a curette, to evacuate the product of the pregnancy.
    
15. The petitioner said that when Abhikesh returned on the Friday evening, Poonam was conscious and able to walk. Unfortunately, Abhikesh refused to take her with him because of the problem at the hostel. This meant that there was no alternative other to allow her to remain in the surgery overnight. When the petitioner was contacted by his receptionist the following morning, he immediately went to the surgery. However, Poonam was by then already dead. The petitioner considered that she had probably died as a result of an air embolism caused by her having moved her legs. This meant that air had been taken into the uterus, and had entered the blood stream.
    
16. The opinion of each assessor was that the petitioner was guilty. The trial judge then delivered reasons for judgment in which he expressed agreement with the assessors, convicted the petitioner, and gave his reasons for doing so.
    

The fresh evidence question

17. Originally, the petitioner confined his challenge to his conviction, in this Court, to the judgment of the Court of Appeal which dismissed his appeal on 23 March 2007. The petition seeking special leave to appeal from that judgment was filed on 2 May 2007.
    
18. However, as recently as 5 September 2008, a second petition was filed. That most recent petition sought special leave to appeal from what was plainly an interlocutory judgment, when the Court of Appeal, on 24 November 2006, refused the petitioner leave to adduce further evidence at the hearing of his appeal.
    
19. This Court has no power to hear and determine any appeal from an interlocutory judgment of the Court of Appeal. Section 122(1) of the Constitution provides that the jurisdiction of this Court is limited to hearing and determining appeals from final judgments of the Court of Appeal.
    
20. Nonetheless, it is well established that where an earlier interlocutory judgment has influenced the outcome of a later final judgment, an appeal against that final judgment may be based on an error in the earlier interlocutory judgment. See generally Crowley v Glissan [1905] HCA 13; (1905) 2 CLR 402 and Gerlach v Clifton Bricks Pty Ltd (2002) 209 CLR 478. The consequence of this is that there was no need for a second petition to be filed. An error on the part of the Court of Appeal in the fresh evidence judgment can form the basis of a successful challenge to the later substantive judgment.
    
21. There is no doubt that the Court of Appeal has power, for good and sufficient reason, to receive fresh evidence, or what is sometimes termed "new" evidence, on appeal. The circumstances in which that power will be exercised have been the subject of numerous statements of principle in the authorities.
    
22. Before this Court, the petitioner relied upon an affidavit sworn by him on 30 July 2008 in support of his challenge to the fresh evidence judgment. That affidavit was sworn to support an extension of time for the filing of the petition relating to that judgment. As no separate petition was required to challenge that judgment, there was no need to seek an extension of time.
    
23. In support of his application to adduce fresh evidence in the Court of Appeal, the petitioner relied upon two previous affidavits sworn by him in October 2006, and February 2007. In those affidavits, the petitioner foreshadowed leading fresh evidence from Dr Andrew Mackinto highly qual qualified specialist gynaecologist resident in New Zealand, and "new" evidence from Dr John Whittaker, a highly qualified gynaecologist also resident in New Zealand who had given evidence for the State at the petitioner’s trial.
    
24. As previously indicated, the State called a number of medical witnesses at the petitioner’s trial. In fact there were five such witnesses, including a forensic pathologist, a toxicologist and Dr Whittaker. The defence did not call any medical witnesses of its own, though the petitioner gave detailed medical evidence in his unsworn statement. It will be recalled that he attributed Po#8217;s death to an air embolism.
    
25. At trial, Dr Whittaker&#s evidence was swas significant. He had not seen Poonam’s body, but was asked to give his opinion from a number of documents and photographs. He was principally asked to say whether, in his opinion, the evidence pointed to a termination, or whether it suggested a miscarriage leading to an evacuation. His conclusion was that there had been a termination, and that it had been carried out quite late in the pregnancy. His opinion was that as a result of the forceful dilation of the cervix, an injury occurred which led to major blood loss. He went on to say that at that point Poonam should have been transferred to a major hospital. However, that did not occur. As a result of the absence of good care from that point onwards, she bled to death. Shock had resulted from major blood loss.
    
26. Dr Whittaker’s evidence was plainly very damaging to the petitioner. His opinions were squarely at odds with the petitioner’s own account of what had occurred. Nonetheless, it was not until after the trial had been eted that the defence sent sent the papers in the case to Dr Macsh seeking his opinion.
27. Dr Mackintosh, in his affidavit, explained that he had been ped with copies of the documdocumentary evidence led at the trial, and also the evidence of ten of the witnesses who testified. These ded all of the medical witnesses. He said that he had discudiscussed the case with Dr Whittaker, he knew, and witd with other specialist obstetricians and gynaecologists. He had also spoken to anaesthetists with experience in this area. Indeed, he said thahad also spoken to a coroner.
    
28. Dr Mackintosh’s concluwaon was that there had been no termination of pregnancy. He said that he considered it more likely that there had been a curette after a partial miscarriage. In his opinion, the most probable cause of death was respiratory failure caused by a number of factors including what he described as "the synergic effects" of the medications given.
    
29. There was also before the Court of Appeal an application to adduce further evidence from Dr Whittaker. He had not at that stage, in the available time, been able to prepare an affidavit. However, he had written a letter setting out the nature of the evidence he now wished to give. That letter commenced:
    

"I have looked at additional written material pertaining to the trial of Dr Mudaliar at which I was employed as an expert witness for the prosecution. This material was not available to me at that time and after discussing the material with Dr Mackintosh I do agree therethere are some areas of uncertainty that were not apparent to me at the time of the trial."

30. The Court of Appeal observed that Dr Whittaker had not explain h in his letter what the "additional written material" to which he had referred consisted of, and had provided no explanation of what he meant by "pertaining t trial". In the remainder of the letter he merely pointed oted out that he was now uncertain, or had doubts about the conclusions he had earlier expressed at the trial.
    
31. The application to adduce further evidence before the Court of Appeal was based upon s 28 of the Court of Appeal Act. That section relevantly provides that the Court of Appeal may, if it thinks it "necessary or expedient in the inteof justice" receive such evidence.
    
32. The Court of Appeal referred to several authorities in dealing with this application. It cited Ladd v Marshall [1954] EWCA Civ 1; [1954] 3 All ER 745 where Lord Denning stated that there were three preconditions to the reception of such evidence on appeal. These were:
    
    • the evidence could not have been obtained prior to the trial by reasonable diligence;
    • it must be such as could have had a substantial influence on the result and
    • it must be apparently credible.
      
33. Their Lordships next referred to Waiske Tuimereke and Anor v State [1998] Cr App AAU 11/97, 14 August 1998 in which the Court of Appeal considered the principles governing the reception of fresh evidence in criminal matters. They referred to Ratten v R [1974] HCA 35; (1974) 131 CLR 510 and Lawless v R [1979] HCA 49; (1979) 142 CLR 659, two of the leading High Court of Australia authorities in which the subject of fresh evidence in criminal matters was canvassed. In both Ratten and Lawless the High Court focussed upon the expression "miscarriage of justice" in the context of intermediate appellate courts dealing with criminal matters.
    
34. In Ratten, Barwick CJ observed that there would be a miscarriage of justice if the jury did not have before it evidence not available to the appellant at the time of his trial which, if believed by the jury, was likely to lead to an acquittal. His Honour added, however, that there would be no miscarriage of justice simply because evidence which was available to the appellant, actually or constructively, was not called, even though it might appear that if that evidence had been called, and believed, a different verdict would most likely have resulted.
    
35. The Court of Appeal in the fresh evidence judgment noted that Barwick CJ’s analysis had been accepted in this country by the Supreme Court in Swadesh Singh v State [2005] CAV 7/05. There the Court stated:
    

"The well established general rule is that fresh evidence will be admitted on appeal if that evidence is properly capable of acceptance, likely to be accepted by the trial court and is so cogent that, in a new trial, it is likely to produce a different verdict ..."

36. Finally, the Court, in the fresh evidence judgment, observed that similar views had been expressed in New Zealand. See generally R v Bain [2003] NZCA 294; [2004] 1 NZLR 638. There, however, Tipping J had added a caution:
    

"Ordinarily if the evidence could, with reasonable diligence, have been called at the trial, it will not qualify as sufficiently fresh. This is not an immutable rule because the overriding criterion is always what course will best serve the interests of justice. The public interest in preserving the finality of jury verdicts means that those accused of crimes must put up their best case at trial and must do so after diligent preparation."

37. The Court of Appeal then applied these principles to the facts of this case. It noted that counsel for the petitioner had acknowledged that other medical evidence, such as that belatedly obtained from Dr Mackintosh, could by reasonable diligence, have been obtained prior to trial. The petitioner had sworn that the decision not to call any such evidence had baken on the advice of his trial counsel, and suggested that that advice had been incompetenpetent.
    
38. The Court of Appeal dealt with that suggestion in the following way:
    

"[24] We must decide whether the evidence is fresh in the sense that it could not have been ascertained prior to the trial with reasonable diligence. With respect to Mr H#8217;s submission, his suis suggestion about the competence of counsel tends to negate his suggestion that it is fresh. The fact it was not obtained because ofdecision of counsel, incompetent or not, demonstrates that that such evidence was available before the trial.

[25] The applicant deposes that he discussed this issue with his counsel a number of times in the period of nearly three years between his arrest and the trial. Counsel’s advice, which he accepted, was that the defence should not call any separate medical evidence and should seek instead to discredit the prosecution evidence by cross examination.

[26] A further hurdle is that the affidavit makes it clear enquiries were in fact made and a report obtained but rejected by counsel because he felt it "was unnecessary". It is also relevant that alternative medical opinion could have been obtained from specialists available at the time in Fiji, two of whom were called by the defence in mitigation.

[27] There could have been a number of reasons why counsel took that course. In the appeal it will no doubt be argued fully and we do not speculate further. However, we bear in mind that defence counsel also had the assistance of his own client’s expertise on all professional medical matters whether or not additional witnesses were called.

[28] The affidavit of Dr Mackintoes not raise any frny fresh matters of fact. It is simply an opinion based on information which was supplied to him and which could equally have been supplied before the trial."

39. In dealing with the further evidence sought to be adduced from Dr Whittaker Court of Appeal real reasoned as follows:
    

[29] Dr Whittaker refers to additional (unspecified) written material. He gives no explanation of the acgrounds for his change of o of opinion; he simply sets out what he suggests are the six components of the two questions he had originally been asked and, in respect of four, states he is uncertain and in another that he no longer holds the same view.

[30] Where any witness has second thoughts about his evidence based on nothing more than a reconsideration of the same information, we cannot accept that satisfies the test of fresh evidence. Mr Gibson for the respondent has told the Court that, should there be a retrial he would not call this witness. That is not surprising as the witness is likely to be considered to have been discredited. The same problem will face the defence if it should seek to call him instead.

[31] In R v Flower [1965] 1 QB 146 it was stated:

"Witnesses may have second thoughts for a variety of reasons. Some become emotionally disturbed, others brood on the effect of their evidence, whilst others are subject to more tangible pressures to induce them to depart from the truth. It is the witness’s state of mind at the trial which matters and this ought to be judged by reference to the circumstances prevailing at the time. It is trite to say that every case depends on its own facts but in our view there is no general requirement for a new trial merely because the witness’s account in this court differs from that given in the court below. So much depends in every case upon the reason, if any, given by the witness for having changed his or her testimony. ... [The witness] gives no acceptable explanation of the reasons for her having changed her story and we feel compelled to reject the evidence."

[32] The position where the witness seeking to change his evidence is an expert involves further considerations. Clearly any expert may consider that recent professional research or newly formed specialist opinion means that his previous opinion is consequently suspect. In such a case, the court would be likely to consider that a sufficient reason for the change and, if the other tests are satisfied, allow it to be introduced to the appeal. But Dr Whittaker’s new evidence does not fall into that category. He does not suggest any new facts or advances in medical knowledge since the trial to persuade him away from this earlier opinion. He has simply developed doubts. That may be understandable in the case of a lay witness but an expert witness is in a special position. He has given his professional opinion and any change will only be credible if it is clearly stated to be based on fresh information earlier about the facts upon which it was founded or in the specialist expertise upon which the determination was based."

40. It was on the basis of these considerations that the Court of Appeal refused the application to adduce the evidence of Dr Mackintosh, or of Dr W160;Whittaker.
    
tli style='text-indent:0pt; margin-top:0pt; margin-bottom:0pt;' value='41' value="41">The petitioner submitted before this Court that the Court of Appeal had approached its task, in dealing whe proposed further evidencidence, incorrectly. He submitted that the fact that Dr Mackintosh&#8 evidence coul could, with reasonable diligence, have been procured before trial was not, as the Court below thought, a virtually insuperable barrier to its recn on appeal. He further submitted that Dr Whittakerer’sence,ence, though presented in a sparse and somewhat unsatisfactory manner, ought to have been regarded, if not as "fresh", then at least as "new", and as warranting consideration in the appear>
42. It is necessary to say a little more about the contents of the proposed fresh evidence. Dr Mackintosh’s affidavit, sworn on 18 October 2006, commenceh a rh a recital of his professional qualifications. They appear to be impressive, and nothing more need be said about them. The no suggestion that he is anything other a highly qualified expert in his field.
    
/li>
43. There is no doubt that Dr Mackintosh had able to him him a great deal of material relating to the death of Poonam. He deposes to having read that material, and to having discussed it with Dr Whit. Dr&Mackintosh said thid that he had also discussed the evhe events surrounding Poonam’s death with a number of other obstetri and gynaecologists in New Zealand, and with an anaesthetist experienced in this general aral area.
    
44. According to Dr Mackintoshpost mortem and tand toxicology information provided showed that there was blood in the uterine cavity. There was some air in the cerebral vessels and cardiac crs, but he did not think that this was a major factor in hain having contributed to the death of the deceased. He noted that she had also had a resolving lung infection.
    
45. Dr Mackintosh commenn the conf confused state of the evidence regarding the gestation of the pregnancy, and the degree of blood loss. He commented that as regards gestation, the observations by Dr Hazratwala werebdominal pall palpation only. The most accurate way of assessing the period of gestation was by ultrasound examination, which had not been carried out. The secost accurate way was by the patient defining her last menstrenstrual period, or how many menstrual periods she had missed. In this case, the evidence was that she had missed two periods. The third most accurate way was by way of abdominal palpation, but this was not necessarily very accurate.
    
46. Dr Mackintosh said that twas lwas little doubt that Poonam had suffered some blood loss at the time of losing her intrauterine contents. He expressed the opinion that some of this blood loss may have occurred at home. Poonam had also given a history of blood loss prior to seeing the Dr Hazratwala. Importantly, twere was no evidence of significant loss of blood in the petitioner’s surgery.
    
47. In dealinh the probable scenario, ano, and the sequence of events leading to the death, Dr Mackintosh expd the opinioninion that, based on Abhikesh’s observation that she had sustained some blood loss before seeing Dr Hazratwala, Poonam had been threatening tcarry even at that stage. The doctor referred to the petitietitioner’s unsworn statement, and concluded that what he had told the assessors was consistent with the external observations made by others. According to the petitioner, Poonam had told him that she had lost the foetus in her toilet, at her home. That meant that the miscarriage was incomplete when she was seen by him in his rooms. Dr Mackintosh sais would be a be a common enough event if the pregnancy were 12 to 14 weeks, as the petitioner claimed. However, it would be unlikely if the foetus were 18 to 20 weeks, as Dr Hazratwala had believed.
    
48. Importantly, Dr Mackinsaid that there wouldwould be some blood loss during any curette carried out by the petitioner. There was blood seen on her undergarments which, from the descri, would be about the amount expected. There was also evidenvidence from Abhikesh supporting this miscarriage thesis. In Dr Macki’s opinion the bthe blood loss theory put forward by the State at the petitioner’s trial was doubtful, far less blooing been detected in the surgery than would have been the case had there been a terminationation, as alleged.
    
49. Dr Mackintosh saat following aing a curette, assuming that that was what had occurred, there appeared to have been some bleeding into the uterine cavity as this was observed bypathologists. Clot retention within the uterine cavity foll following a curette would not be uncommon. It could cause the uterus to enlarge, and bring about haemorrhage. The blood would be retained and would not seep through the cervix, into the vagina. This too appeared to be consistent with the observations made by the pathologist.
    
50. Commenting on the medication administered by the petitioner, Dr Mackintosh said this would ould produce respiratory depression, and a synergic reaction. If Poonam had been recovering from a chest infection (here was some evidence to suggest that she had a cold at the time) it could well be that that the cause of death was, in fact, respiratory failure, primarily caused by the synergic effect of the medication administered. He said that the medication used was "rather antiquated" and, although very effective, no longer used in New Zealand because of the risk of respiratory failure, and the possibility of a marked synergic effect. He commented that based on her appearance, Poonam could well have been anaemic. That fact, coupled with the blood loss that she had sustained as a result of the curette, the possibility that she may have been recovering from a cold, and the effects of the medication given to her, could well have resulted in respiratory failure which led to death.
    
51. It is unnecessary to spell out in any further detail the contents of Dr Mackintosh&;s affidavit. Int. In short, he was highly critical of the evidence given by the various medical witnesses called by the State at the trial. He rejected entirely the proposition that the pregnhad been of 20 weeks gestatestation. He explained that Poonam, having been a thin Indian woman of probably less than 50 kilograms in weight, was more likely to have been about 12 to 14 weeks pregnant. The assumption that she died from blood loss could not be sustained. A much more likely scenario was that she did not lose anything like the two or three litres of blood that would have been necessary to result in death from this cause, but rather significantly less than one litre. Had there been massive blood loss, the petitioner would certainly have needed to clean up the surgery on the Friday evening, after Abhikesh had left. There was nothing to suggest that he had done so. Moreover, there were no instruments found by the police in the petitioner’s surgery capable of having been used to carry out a termination of a 20 week old foetus.
    
52. In conclusion, Dr Mackintosh reiterated it wait was his belief that the petitioner had not carried out an abortion. It was more likely than not, in his view, that he had simply performed a curette aftonam had partially miscarriaged at home.
    
53. As previously indicated, Dr Whittaker did not swn affidaffidavit. However, by letter dated 9 October 2008, addressed to the petitioner’s senior counsel in Auckland, he reported that he had examined additional material pertaining to thitioner’s trial. He s He said that this material had not been available to him at the time. After discussing the matter with Dr Mackintosh, he agreed thatethere were "some areas of uncertainty" not apparent to him at the time of the trial.
    
54. Having previously sworn at the petitioner’s trial that it was his opinion that the petitioner had carried out a termination, he now said that he was uncertain, "and even doubtful" that this was so. He accepted that she had not been 18 to 20 weeks pregnant, as Dr Hazratwala had said, but more likely, based on two missed menstrual periods, about 12 weeks pregnant. He now accepted that the cervical tear that had previously caused him to conclude that there had been an abortion may have been caused by a curette. In answer to the question whether the patient should have been transferred from the petitioner’s rooms, in the absence of visible bleeding, he now said that he could no longer state that this would be expected, at least in Fiji, at the time of this incident. Finally, he resiled from his earlier conclusion that Poonam had bled to death. He now said that bleeding had only been one of several factors that contributed to her demise.
    
55. The Court of Appeal considered both the affidavit and the letter, but determined that neither gave rise to any basis for calling Dr Macsh, or Dr Whittakertaker on the l. Inl. In arriving at that conclusion, their Lordships said that they were greatly influeny the choice that the petitioner had made, on the advice of his trial counsel, not to seek seek out and obtain medical evidence to support his case. In that regard the petitioner had adopted a particular course. He had to bear the consequences of having done so.
    
56. In addition, the Court of Appeal regarded the letter prepared by Dr Whittaker as wholly inade, ate, in the circumstances, to form a basis for granting leave to adduce further evidence from him.
    
57. The Court of Appeal, hav, having considered mainly Australian authority on the subject of fresh evidence, might have referred to High Court decisions more recent than Ratten and Lawless. The cases now firmly establish that at common law, a distinction is drawn between "fresh" evidence, that is to say evidence which was not known to the appellant at the time of his trial, and which could not have been discovered by him with reasonable diligence, and evidence which is merely "new". See for example Gallagher v The Queen [1986] HCA 26; (1986) 160 CLR 392 at 402 and 410; and Mickelberg v The Queen [1989] HCA 35; (1989) 167 CLR 259 at 301. Evidence which was either available, or could with reasonable diligence have been discovered, before trial is not "fresh" evidence.
    
58. In Mallard v The Queen [2003] WASCA 296; (2003) 28 WAR 1 the Western Australian Court of Appeal discussed the application of this distinction in criminal cases. The Court observed (at [14]):
    

"... as was noted in Ratten, the accused person in the case of a criminal trial is afforded considerable latitude, because of the difficulty of the accused's position and the discrepancy between the resources perceived to be available to the Crown and to the accused. Evidence not actually available to an accused will often be treated as fresh evidence, notwithstanding that it could on a narrow view have been discovered by diligent inquiry. That is something which falls to be evaluated having regard to the circumstances of each case."

59. Strictly speaking, if evidence is not "fresh" but merely "new" an appellate court will quash a guilty verdict only if that evidence either shows the appellant to be innocent, or raises such a doubt about his or her guilt in the mind of the court that the verdict should not be allowed to stand.
    
60. A less stringent threshold applies to "fresh" evidence. As Gallagher explains, where there is "fresh" evidence, an appellate court will conclude that the trial involved a miscarriage of justice if the court considers that there is a significant possibility that the jury, acting reasonably, would have acquitted the appellant if that evidence had been before it at the trial: at 399 per Gibbs CJ and at 402 per Mason and Deane JJ.
    
61. Plainly, a decision deliberately taken by an accused not to adduce evidence of a particular kind at trial will weigh heavily against its reception on appeal. However, no invariable rule concerning the failure to call such evidence can or should be laid down. The discretion conferred upon the Court must be exercised judicially, but having due regard to the interests of justice, above all else.
    
62. Having given this matter careful consideration, we are not persuaded that the Court of Appeal erred, in the exercise of its discretion, in rejecting the petitioner’s application to adduce the evidence of Dr Mackintosh, and of Dr #160;Whittaker, in the sutitantive appeal.
    

The original petition

63. The Court of Appeal, in its subsve judgment, grouped the grhe grounds of appeal into five main topics. The first point taken was that the indictment was bad for duplicity. The Court of Appeal rejected that ground, and an associated complaint of latent uncertainty. Although these points were taken again in the petition to this Court, Mr Haigh QC, on b of the petitpetitioner, conceded that any problems there may have been with regard to these matters had been overcome by the trial judge’s reasons for judgment. The ground was not pressfore us, and nothing more nore need be said about it.
    
64. The next matter raised before the Court of Appeal, and repeated in the petition to this Court, concerned the failure on the part of the trial judge to provide to the assessors an accomplice warning. The petitioner’s case is, and always has been, that Abhikesh was an accomplice, and that the trial judge was therefore obliged to have warned the assessors of the dangers of convicting on his evidence, in the absence of corroboration. The State has always taken the position that Abhikesh was not an accomplice, and that no such warning was required.
    
65. The Court of Appeal observed, correctly, that the evidence of Abhikesh was of central importance to the prosecution case. It was Abhikesh who described what had occurred when Poonam saw Dr Hazratwala. Dr Hazratwala had l, if any, rny, recollection of having seen Poonam, or examined her. It was Abhikesh who said that Dr Hazratwala had stated m wast 20 weeks pregnant. It was Abhikesh who described what had occurred at the petitpetitionerioner’s surgery on 19 March 2003, and again two days later. It was he who described Poonam’s condition as being consistent with the prosecution case, and in direct contrast to the petitioner’s account. It was Abhikesh who gave evidence that he had withdrawn the sum of $950 from the bank to pay for what he said was a termination, though as the trial judge properly pointed out to the assessors, that sum could just as well have been required to meet the expenses incurred through the performance of a curette.
    
66. It was no doubt in order to make the evidence of Abhikesh available to the State that he was granted total immunity from prosecution. He could of course, on his own admission, have been charged with counselling or procuring, or aiding or abetting, an unlawful abortion.
    
67. The Court of Appeal concluded that Abhikesh was not an accomplice within the meaning of that term as explained by the House of Lords in Davies v DPP (1954) 38 C App R at 11. Abhikesh was not particeps criminis in the offence of manslaughter. He was only involved, if at all, in the carrying out of the abortion. As such, no warning as to the dangers of convicting upon his evidence in the absence of corroboration was necessary.
    
68. In arriving at that conclusion the Court of Appeal had regard to s 21 of the Penal Ci>. That section could, in theory, have rendered Abhikesh an accomplice to the charge of manslaughter. However, as the Court of Appeal noted, it would be difficult to accept that an aon performed by an experienerienced gynaecologist, such as the petitioner, would "probably" have led to Poonam’s death. Accordingly, so the Court of Appeal held, Abhikesh could not properly have been regarded as an accomplice to manslaughter.
    
69. Having arrived at that conclusion, the Court of Appeal went on to observe that the trial judge had still acted correctly in directing the assessors to scrutinise the evidence of Abhikesh with care. His Lordship had told them to approach that evidence "with caution". He had directed them to look for items of evidence independent of the evidence of Abhikesh that would confirm parts of his account. Regrettably, in their Lordship’s view, even after giving these directions, the trial judge had then watered them down, to some degree, by adding:
    

"Although it is not entirely clear how his position would be improved by saying that it was an abortion procedure rather than an attendance to a miscarriage, it is possible he may have wished to please the police rather than to tell the truth."

70. The trial judge did remind the assessors that Abhikesh had been granted immunity from prosecution. He told them that this related to his possibly having been implicated in the abortion itself. What he failed to do was to explain to the assessors precisely why Abhikesh’s evidence may have been tainted by an improper motive. That is a fundamental aspect of any accomplice warning, but it applies with equal force to those cases in which, though technically an accomplice warning is not required, a warning closely analogous thereto should be given.
    
71. The matter is dealt with at some length in Archbold Criminal Pleading Evidence and Practice 2007 at par 4-404n. In R v Beck [1982] 1 WLR 461 Ackner LJ giving the judgment of the Court of Appeal, referred to "the obligation upon a judge to advise a jury to proceed with caution where there is material to suggest that a witness’s evidence may be tainted by an improper motive". Beck has been repeatedly applied in England. In Chan Wei-keung v R [1995] 2 Cr App 194 the Privy Council stressed that where a witness has potential ulterior motives, such as the hope of obtaining a discount in proceedings against himself, it is important that those ulterior motives should be put squarely before the jury. Recently, a similar approach was taken in Benedetto v R [2003] EWHC 174; [2003] 1 WLR 1545, dealing with confessions made to fellow prisoners.
    
72. It is not sufficient for a trial judge simply to warn assessors of the need for caution before acting on disputed evidence. As will become clear Abhikesh had a strong incentive to make his evidence of what had occurred conform to the prosecution theory that Poonam had attended the petitioner’s surgery for an abortion. His initial account was, of course, quite different, and accorded with the petitioner’s defence. He later changed his story, and was granted immunity. The warning given by the trial judge in this case fell well short of what was required, and the Court of Appeal recognised that fact. Indeed, their Lordships described the trial judge’s directions to the assessors as "sparse to the stage of inadequacy".
    
73. The Court of Appeal also noted that the trial judge had incorrectly told the assessors that a condition of the immunity granted to Abhikesh was that he tell the truth. Rather, the immunity was couched in the usual terms, namely it was given on the basis that Abhikesh would give evidence for the State in accordance with an earlier statement he had provided. That statement implicated the petitioner.
    
74. Of particular importance, the conditional nature of the immunity granted to Abhikesh carried the risk that he would adhere to his earlier statement to the police, whether or not it was true, because of the fear that he would face prosecution if he did not.
    
75. To make matters worse, the trial judge directed the assessors that the petitioner’s counsel had not suggested to Abhikesh that he had been pressured by the police. That direction was, unfortunately, incorrect.
    
76. Despite these significant errors on the part of the trial judge, the Court of Appeal ultimately concluded that the failure to give the assessors appropriate warnings regarding the evidence of Abhikesh had not led to a miscarriage of justice. They did not invoke the proviso.
    
77. Their Lordships reasoned as follows. Trial counsel for the petitioner had failed to take any exception to the trial judge’s directions regarding the way in which the assessors should approach Abhikesh’s evidence. Counsel for the State had raised that issue prior to the closing addresses, and submitted that no accomplice warning should be given. Counsel for the petitioner said that he had nothing to add. At the conclusion of the trial judge’s summing up, counsel for the petitioner took a number of exceptions, but said nothing about the failure to give an adequate warning.
    
78. The Court of Appeal was obviously suspicious of all this. The record showed that, immediately after sentence, when trial counsel sought bail pending an appeal, he submitted that the failure to give an appropriate warning would be one of the grounds that would be argued in any such appeal. This suggested to their Lordships that trial counsel had deliberately kept the point for appeal, rather than raising it to ensure a fair trial.
    
79. With respect, this seems a somewhat harsh finding. There was a significant time lag between the date on which the accomplice issue was raised by counsel for the State, on 25 April 20nd the date on whic which the petitioner was sentenced, on 17 Ma6. His Lordship summed uped up to the assessors on 15 May 2006. possible that trat trounsel failed to take exception to the summing up, not as a as a deliberate ploy, but simply in ignorance of the law on this subject.
    
80. The earlier reference to the need for a warning had been made almost in passing, and was dealt with in less than one page of transcript. The main point of State counsel’s submission was to persuade the trial judge that Abhikesh was not an accomplice, and that the fact he had been granted immunity should not attract a full corroboration warning. It may be that trial counsel finally realised, by the time that he made the bail application on behalf of his client, that the warning had in fact been deficient.
    
81. In any event, if the failure to give an appropriate direction in relation to Abhikesh’s evidence led to a possible error on the part of the assessors, and indeed, on the part of his Lordship as well, because it failed to bring out the dangers associated with acting upon that evidence, it would take a good deal to convince us that failure to take the exception should be regarded as fatal in this case. Abhikesh was fundamental to the prosecution case. His credibility was of critical importance. The account that he gave differed starkly from that of the petitioner. That suggests that the petitioner may have suffered serious prejudice by reason of a warning that was, in the circumstances, quite inadequate.
    
82. The failure to give an adequate warning regarding corroboration was not the only error that the Court of Appeal found in the trial judge’s summing up to the assessors. Counsel for the petitioner raised a number of matters, including defects in the directions regarding the burden of proof, which the Court of Appeal accepted as having been established. Some of these defects might be seen as being of greater significance than others. In any event, the Court of Appeal concluded that none of this mattered. Though it described these various misdirections as "unfortunate", the Court of Appeal concluded that the trial judge had "cured the problem" because he had directed himself correctly in his reasons for judgment, once the assessors had stated their opinions.
    
83. We shall return to whether that was a correct approach for the Court of Appeal to have adopted shortly. It is first necessary to identify some of the specific errors which the Court of Appeal concluded had been made.
    
84. Importantly, their Lordships held that the trial judge had failed adequately to summarise the evidence to the assessors. The summary was of particular importance in this case because of the unfortunate fact that, at the conclusion of counsels’ closing addresses, the Court adjourned for eighteen days before his Lordship summed up. It was clear that the judge had arranged to go abroad, and the trial clearly took longer than anticipated. The Court of Appeal described this delay as "unfortunate" in a trial with assessors. With respect that seems an understatement.
    
85. The Court of Appeal concluded that the trial judge had not adequately put the defence case to the assessors. Apart from cross-examination, that case was presented entirely in the petitioner’s own statement from the dock. Their Lordships observed at par [55] of their reasons for judgment that in such a case "the judge should always ensure that the accused’s statement is put clearly to the assessors, if necessary by reading it to them." They observed that the medical evidence in this case was of critical importance. The petitioner had given a detailed account of various medical matters in his unsworn statement which, incidentally, ran for some thirteen pages of single space text in the record of the Court. Although some weeks had passed from the time the unsworn statement was made to the date of the summing up, the trial judge did not remind the assessors of what the petitioner had said save in one or two brief paragraphs. He made no mention whatever of the medical matters to which the petitioner had directed much attention. This led the Court of Appeal to observe that "it would undoubtedly have been better if the judge had repeated and explained their significance".
    
86. By contrast with the way in which the trial judge summarised the defence case, the prosecution was said, by the Court of Appeal, to have "fared better". Having first reminded the assessors of the evidence given by Abhikesh and by Dr Hazratwala, his Lordship set out in considerable detail a summary of the evidence given by the prosecution experts which ran, in transcript to five and a half pages of text. Following that there was no analysis of the tecl aspects of the petitionerioner’s account.
    
87. Overall, their Lordships concluded that whatever the trial judge had said to the assessors, it "clearly did not amount to a fair summary of the appellant’s medical explanation".
    
88. The Court of Appeal found other defects as well. It was contended on the part of the petitioner, and accepted by their Lordships, that the trial judge gave an incorrect account to the assessors of the legal effect of s 234 of the Penal Code. That section provides:
    

"A person is not criminally responsible for performing in good faith with reasonable care and skill a surgical operation upon any person for his benefit, or upon an unborn child for the preservation of the mother’s life, if the performance of the operation is reasonable, having regard to the patient’s state at the time and to all the circumstances of the case."

89. The trial judge told the assessors, regarding that section, that
    

"It is not part of the defence case that the accused made any attempt to bring on a miscarriage or to carry out an abortion, so you are not concerned to consider whether such operations are done in good faith or in the interests of the health and life of Poonam."

90. The Court of Appeal found that this direction was incorrect. The defence case was that Poonam had already miscarried and so the procedure that the petitioner performed was simply a curette. That was a surgical procedure for the benefit of Poonam, and therefore a matter to which s 234 was rele His Lordship ship should have directed the assessors on the need to consider that, if what was performed was a curette, and notbortion, the section might apply if the acts done were carried out in good faith, and were were reasonable. In the language of the Court of Appeal:
    

"Those were clearly matters which should have been left for the assessors to consider and the manner in which the judge dealt with them amounted to a misdirection."

91. There were other points raised, but it is unnecessary to address them here. It is sufficient to say simply that the Court of Appeal expressed a number of concerns about his Lordship’s directions. Their Lordships said at par [67]:
    

"We accept that the form of the summing up and the individual directions give cause for disquiet."

92. One might have thought that in the light of this observation the Court would have gone on to consider whether, notwithstanding the errors that the trial judge had made, the proviso could be invoked to save this conviction. However, the Court did not approach the matter in that way. Rather their Lordships said there were two answers to the petitioner’s case.
    
93. The first was the procedure under Fijian law whereby the judge has the sole responsibility to determine the verdict. After referring to s 2 the Criminal Procedrocedure Code their Lordships noted that in the present case the trial judge had concurred with the assessors’ opinions and did not, under s 299 ha provny reasons beyonbeyondeyond his decision. However, his Lordship had done so and, in those reasons, explained how he reached the ct that he had. According to the Court of Appeal:
    

"Many of the problems with the summing up are cured by his reasoning in his judgment and we consider they must be taken by this Court as the basis for the ultimate finding of guilt."

94. The second factor that led the Court of Appeal to find that there had been no miscarriage of justice was essentially as follows. However the assessors may have been directed, the unchallenged evidence of the manner in which a qualified doctor left an indisputably seriously injured woman alone, who died as a result, as the trial judge found, was sufficient in itself to justify a conviction of manslaughter.
    

Disposition of the petition

95. We have already indicated that we are not persuaded that the Court of Appeal, in the fresh evidence judgment, erred in its refusal to allow the evidence of Dr Mackintosh, and that of160;#160;Whittaker to be adduced on appeal.
    
96. The decision of the Court in the substantive appeal is an altogether different matt matter. Section 7(2) of the Supreme Court Act 1998 requires this Court to be satisfied that one or more of the conditions set out therein is made out before special leave can be granted. In our view, this application raises a question of general legal importance, as well as a substantial question of principle affecting the administration of criminal justice.
    
97. The point of principle involved is whether the Court of Appeal correctly approached its task when considering whether there had been a miscarriage of justice. Put simply, it determined that whatever errors the trial judge may have made in his summing up to the assessors, these were of no consequence. That was because the ultimate decision that the petitioner was guilty was made by his Lordship, and not by the assessors, and in his reasons for judgment the errors were "cured".
    
98. At one level that is a surprising proposition. Both the Court of Appeal, and this Court have traditionally focussed upon the directions given by trial judges to assessors, when considering whether to permit a conviction to stand. If the approach taken by the Court of Appeal in this case were to be followed generally, it would scarcely be necessary to have regard to the adequacy, or otherwise, of a summing up. Any deficiencies, even serious ones, could be "cured" if the trial judge elected to give reasons, and indicated that he had directed himself or herself correctly.
    
99. This seems quite wrong in principle. At common law the position is clear. An accused person is entitled to a trial in which the relevant law is correctly explained to the jury and the rules of procedure and evidence are strictly followed. In Mraz v R [1955] HCA 59; (1955) 93 CLR 493 Fullagar J said at 514:
    

"If there is any failure in any of these respects, and the appellant may thereby have lost a chance which was fairly open to him of being acquitted, there is, in the eye of the law, a miscarriage of justice. Justice has miscarried in such cases, because the appellant has not had what the law says that he shall have, and justice is justice according to law."

100. Not only does the approach taken by the Court of Appeal in this case seem contrary to principle, it also runs directly contrary to authority. We refer in particular to Baharat v The Queen [1959] AC 555 in which the Privy Council delivered a judgment on appeal from the Court of Appeal of Fiji.
     
101. The facts of Baharat were as follows. At the trial of the appellant on a charge of murder, at which he gave evidence which raised issues of self-defence and provocation, the judge wrongly directed the assessors that on the issue of provocation the burden of proof rested upon the accused. He also misdirected the assessors as to what amounted in law to provocation. Each assessor gave his opinion that the appellant was guilty of murder. The judge subsequently gave judgment to the same effect. The Court of Appeal of Fiji dismissed the appellant’s appeal against his conviction.
     
102. The advice of the Privy Council was given by Lord Denning who said at 539-40:
     

"What is the consequence of the misdirection given by the judge to the assessors? According to section 246 of the Criminal Procedure Code the trial is by the judge "with the aid of "assessors"." The judge is not bound to conform to their opinions, but he must at least take them into account. If they have been misdirected on a vital point, their opinions are vitiated. Take this very case. Suppose the assessors had been properly directed, is it not possible that one or more of them might have been of the opinion that the appellant was guilty of manslaughter only? If the majority of them had given such an opinion, the judge might possibly have accepted it in preference to his own. At any rate he could hardly have rejected it without saying why he did so. He has, in truth, by his misdirection, disabled the assessors from giving him the aid which they should have given; and thus in turn disabled himself from taking their opinions into account as he should have done. This is a fatal flaw.

On behalf of the Crown it was said that the misdirection did not matter, because the judge disbelieved the appellant’s story altogether, and there was thus no foundation on which an issue of provocation could be raised. Their Lordships cannot accept this contention. Apart altogether from the appellant’s evidence, there was some evidence of a struggle having taken place. Their Lordships think this was clearly a case where the question of provocation should have been considered: see Bullard v The Queen [1957] AC 635; [1957] 3 WLR 656. The failure of the judge to direct the assessors properly upon it, or to consider it himself in his summing-up, means that the judgment cannot stand."

103. It is difficult to reconcile the reasoning of their Lordships in Baharat with that of the Court of Appeal in this case. In Baharat the assessors and the trial judge were misdirected. That same is true in the present case. Indeed, the trial judge made that plain. At the commencement of his reasons for judgment he stated the following "I direct myself in accordance with my summing up." He did not, in any way, correct his misdirection to the assessors regarding the evidence of Abhikesh. As we have said, something close to an accomplice warning was required in this case and his Lordship failed to provide it.
     
104. The same could be said of the trial judge’s failure adequately to put the defence case to the assessors, by reason of the somewhat cursory treatment given to the petitioner’s unsworn statement in the summing up. We need hardly cite authority for the proposition that a judge must ensure that the defence case is adequately presented. The cases are set out in Archbold at par 4-36.
     
105. Mr Elliott, who appeared before this Court on behalf of the State, conceded very fairly, that on its face, Baharat appeared to run directly counter to the view taken in the case by the Court of Appeal that defects in the summing up could be cured by the ultimate judgment of the trial judge. He accepted that the Privy Council had clearly articulated the very point which concerned this Court, in argument, that by reason of his own misdirection, the trial judge may not have obtained the properly formed opinions of the assessors.
     
106. Mr Elliott submitted, however, that this was not the end of the matter. He submitted that whilst it was best that there be no misdirection when summing up to the assessors, the question in the present case was whether, the judgment being that of the judge, a series of misdirections to the assessors could be cured if the reasons supporting the verdict were expressed correctly.
     
107. Mr Elliott sought to distinguish Baharat by arguing that this particular point had not arisen for determination in that case. That was because the judge in Baharat, despite having given a reasoned judgment, still did not deal with the question of provocation, upon which he had plainly misdirected the assessors. It followed that the judge and the assessors were both misdirected in that case.
     
108. According to Mr Elliott that was a sufficient basis for allowing the appeal in Baharat. The Privy Council need not have gone further. Their Lordships had, however, done so, considering the effect of s 246 of the Criminal Procedure Code and holding that a misdirection to the assessors disables the judge from taking their opinions into account as required by that sec Put simply, Mr Elliott submittet what thei their Lord Lordships had decided in Baharat regarding this wider point was dicta, and not form any part of the ratio of the case.
     
109. We are not persuaded by that submission. The reasoning of the Privy Council, which underpinned its decision, had as its most critical feature the passages regarding the disabling of the assessors, and the disabling of the judge, to which we have referred. Of course this Court is not bound by decisions of the Privy Council. However, a considered judgment by that Court, on appeal from Fiji, would not lightly be ignored.
     
110. Mr Elliott had a second string to his bow. He noted that s 246 of the Criminal Procedure Code was repealed in November 1961. The original language of the section required trials before the Supremrt to be "with the aid of assessors in lieu of a jury". That language was not replicated ined in s 299(1) of the Criminal Procedure Code which simply requires the judge to sum up, and then ask each of the assessors to state his opinion orally>. Section 299(2) provides that the judge shall then give judgment, but in doing song so shall not be bound to conform to the opinions of the assessors. The sub-section goes on to provide that where the judge’s summing up of the evidence under sub-section (1) is on record, it shall not be necessary for any judgment, other than the decision of the court, which shall be written down, to be given. The exception is that when the judge does not agree with the majority opinion of the assessors he must give his reasons for differing with that opinion, and in every such case the judge’s summing up and the decision of the court, together with, where appropriate, the judge’s reasons for differing with the majority opinion of the assessors, is deemed to be the judgment of the court.
     
111. Mr Elliott submitted that s 299 was on the statute bat s at the time Baharat was decided as s 281 and t from the proviso viso in sub-section (2)) in substantially the same form as it now stands. He submitted that the language&#160leads intractablytably to the conclusion that in Fiji the judge, quite separately from the the assessors, decides the case and the role of the assessors is nothing more than advisory. He therefore submitted that Baharat is of no assistance to the petitioner. In his own words, "unsatisfactory as it may seem s 299 requires an appellate court to disregard defects in a summing up if the judge’s reasons make it clear that the judge has applied the law correctly.
     
112. In our view this variant of the argument should also be rejected. Despite the repeal of s 246, assessors continue to play a vital role in the administration of justice in Fiji. That role was considered by the Privy Council in a case from Swaziland, Maklikilili Dhalamini & Ors v The King [ AC 583. Their Lordships obps observed at 588 that the giving of opinions by those who were, for practical purposes "assessors" was part of the proceedings of the High Court, and had to be carried out in open court.
     
113. The Privy Council went further and stated at 589-90:
     

"It must, further, be remembered that the provision of giving the judge, at his request, the assistance of a native assessor cannot be regarded solely from the point of view of aid given to the judge. It operates, and no doubt is intended to operate, as a safeguard to natives accused of crime, and a guarantee to the native population that their own customs and habits of life are not misunderstood. From this point of view the importance of publicity is manifest. Under the present practice, judging by the present record, it is not made known to the public whether any opinion was given at all by the assessor, or what it was. The High Commissioner alone is informed in the judge’s report. So far is it kept secret that it was not until this appeal was opened and a special request for information addressed by their Lordships to the Protectorate that the facts were made known for the purpose of appeal. There seems, therefore, no reason for refusing to give to the sections in the proclamation the meaning which the words clearly indicate.

What, then, should be the result of the failure to comply with the proclamation and to hold the whole of the proceedings in public? In this country the omission would be a fatal flaw entitling a convicted criminal to have the conviction set aside. An analogous case is where the judge has either pronounced sentence or altered sentence in the absence of the accused: see Lawrence v The King [1933] AC 699, where a judge in Nigeria had altered sentence both in the absence of the accused and when sitting in chambers. Prima facie, the failure to hold the whole of the proceedings in public must amount to such a disregard of the forms of justice as to lead to substantial and grave injustice within the rule adopted by this Board in dealing with criminal appeals. There may, no doubt, be cases where the guilt of the accused is so apparent that in spite of the disregard of this essential need for publicity this Board would not consider it right to grant leave to appeal, but the present is not such a case, as a particular native custom formed an important consideration on which it was essential that the proclaimed necessity for publicity should be observed. For these reasons their Lordships came to the conclusion that they should recommend to His Majesty that the appeals should be allowed."

114. These observations remain apposite to Fiji today.
     
115. Had the Court of Appeal had its attention drawn to Baharat, it would not, in our view, have approached this case in the way that it did. Errors in summing up to assessors cannot be overlooked simply because the trial judge has gone on to give reasons for decision in which those errors are not replicated. It would be contrary to all justice if a summing up containing serious errors could be largely ignored by the Court of Appeal provided that the trial judge, in his reasons for judgment, given after the assessors had expressed their opinions, happened to state the law correctly. Public confidence in the administration of criminal justice in this country would be diminished if assessors were relegated to such an inconsequential role.
     
116. In the present case, the trial judge:
     
     • misdirected the assessors on significant issues which left the Court of Appeal uneasy, and concerned; and
       
     • stated in his own reasons for judgment that he was directing himself in accordance with the summing up, delivered the previous day, and which contained, as the Court of Appeal itself found, a number of significant errors.
       
117. Notwithstanding the conclusion of the Court of Appeal, recorded at par [68] of its substantive judgment, it is incorrect to say that the problems revealed in the summing up were "cured" by the trial judge’s reasons for judgment.
     
118. In our opinion, the Court of Appeal erred in allowing the petitioner’s conviction to stand. We say that fully cognisant of the second factor that influenced that Court in arriving at its decision. Whether there was indeed gross negligence on the part of the petitioner, and whether that negligence led to Poonam’s death, were questions that could not be considered in isolation. The credibility of Abhikesh was critical to those issues. The trial judge found that Abhikesh was a truthful and reliable witness. However, he did so having failed to direct himself properly as to how the evidence of that witness should be approached.
     
119. Having regard to the errors in the summing up that were identified by the Court of Appeal, with whose judgement we agree in that respect, we consider that there was a miscarriage of justice in this case. The Court of Appeal ought to have ordered that there be a new trial.
     
120. We note that the petitioner has served his sentence in its entirety. We understand that a key witness for the prosecution would no longer be available to give evidence if there were a retrial. In addition, the State would have to contend with the evidence of Dr Mackintosh and Dr Whittaker.ough we have reje rejected the contention that their evidence ought to have been received by the Court of Appeal as fresh evidence, we are cognisant of its possible effect on a retrial of mattbr>
121. So too was Mr Ellioho fairly conceded thad that if there were to be a retrial, he would urge that it be confined to a case of manslaughter by grossigence. That of course was not the basis upon which the trial in this matter was conducted.cted. Gross negligence, though left to the assessors as an alternative basis upon which they could find manslaughter, was very much a secondary aspect of the prosecution case.
     
122. Whether there is in fact to be a retrial must be a matter for the Director of Public Prosecutions. We can do no more than suggest that the most careful consideration be given to whether the public interest requires that a retrial be held.
     
123. The orders of the Court are as follows:
     
     1. Special leave to appeal be granted.
     2. The appeal be allowed and the judgment of the Court of Appeal be set aside.
     3. The conviction on the count of manslaughter be quashed and the sentence set aside.
     4. There be a new trial, at the election of the Director of Public Prosecutions.
        

Hon Justice Keith Mason
Judge of the Supreme Court

Hon Justice Kenneth Handley
Judge of the Supreme Court

Hon Justice Mark Weinberg
Judge of the Supreme Court

Solicitors:

Naidu Law, for the Petitioner
Office of the Director of Public Prosecutions, Suva for the Respondent