The Judicial Committee of the Privy Council Decisions

Barlow v. The Queen (New Zealand) [2009] UKPC 30 (08 July 2009)

Privy Council Appeal No 32 of 2008

John Robert Barlow Appellant

v.

The Queen Respondent

FROM

THE COURT OF APPEAL OF

NEW ZEALAND

- - - - - - - - - - - - - - - - -

JUDGMENT OF THE LORDS OF THE JUDICIAL

COMMITTEE OF THE PRIVY COUNCIL

Delivered the 8^th July 2009

- - - - - - - - - - - - - - - - -

Present at the hearing:-

Lord Scott of Foscote

Lord Rodger of Earlsferry

Lord Walker of Gestingthorpe

Lord Neuberger of Abbotsbury

Sir Christopher Rose

- - - - - - - - - - - - - - - -

[Delivered by Lord Scott of Foscote and Lord Rodger of Earlsferry]

Introduction

1. In November 1995 John Barlow was found guilty of murdering Eugene Thomas and his son Gene Thomas at Wellington on 16 February 1994 and was sentenced to life imprisonment. This was the third trial on that charge: two earlier trials in 1995 had ended with the jury unable to reach agreement after a combined total of five and a half days of deliberations. The third jury reached their verdict after 27 hours of deliberation. Mr Barlow appealed, but on 21 August 1996 the Court of Appeal dismissed his appeal. Mr. Barlow remains in prison.
2. On 22 July 2008 the Judicial Committee heard a petition by Mr. Barlow for special leave to appeal to the Privy Council against the 1996 decision of the Court of Appeal to dismiss his appeal. The petition was based on new evidence which had become known and which their Lordships will later describe.
3. The Crown opposed the petition on two grounds: first, on the merits, it was submitted that the new evidence did not render the conviction unsafe. Secondly, the Crown took a jurisdiction point. On the basis of the new evidence Mr Barlow had asked the Governor-General, pursuant to section 406 of the Crimes Act 1961, to refer the question of his conviction back to the Court of Appeal. The power of the Governor-General to do so is a statutory addition to the Governor-General's power to exercise the prerogative of mercy. Section 406 provides that

"Nothing in this Act shall affect the prerogative of mercy, but the Governor-General in Council, on the consideration of any application for the exercise of the mercy of the Crown …, may … if he thinks fit … either

(a) Refer the question of the conviction …

to the Court of Appeal …. or

(b) If he desires the assistance of the Court of Appeal on any point arising in the case with a view to the determination of the application, refer that point to the Court of Appeal for its opinion thereon …."

4. The Minister of Justice is the constitutional adviser of the Governor-General in relation to applications under section 406 for the exercise of the prerogative of mercy and the Minister, in turn, is advised by officials in the Ministry with the responsibility of inquiring into, and reporting on, any such application. It is, their Lordships were told, the invariable practice of the Ministry to ask senior external counsel either to report to the Minister or to review the proposed report prepared by the departmental officials. In the case of Mr Barlow's application a Ministry report was prepared and then reviewed by Dr Fisher QC, a retired High Court judge. Dr Fisher, after reviewing the proposed report and the Ministry file, agreed with the conclusion of the report that

"The applicant has not produced relevant and cogent evidence pointing to a likely miscarriage of justice, such that it is appropriate to either grant a free pardon or to refer the question of his convictions to the Court of Appeal."

So Mr Barlow's section 406 application was dismissed.

5. In Thomas v The Queen [1980] AC 125, 136, Lord Edmund Davies, giving the judgment of the Board, held that, where the Governor-General had referred a point to the Court of Appeal under section 406(b), the Court of Appeal's opinion on the point could not be made the subject of an appeal to the Privy Council:

"The opinion [the Court of Appeal] expressed impinged upon no legal right of the defendant, nor did it place any fetter upon the exercise by the Governor-General of the royal prerogative of mercy."

It follows that the Governor-General's dismissal of Mr Barlow's application could not have been the subject of any appeal. The Crown has submitted that Mr Barlow's present application for leave to appeal to the Board is, in substance although not in form, an attempt to reverse the Governor-General's decision to dismiss his section 406 prerogative of mercy application and, accordingly, that the Board has no jurisdiction to entertain the application.

6. Bearing in mind the potential constitutional significance of this jurisdiction objection to Mr Barlow's leave application to the Board, the Committee of three that heard the application directed that it be adjourned to come on before a Committee of five on the footing that, if leave were granted, the hearing of the appeal would immediately follow. This judgment of the Board is given after the hearing of Mr Barlow's adjourned application for leave to appeal against the Court of Appeal's 1996 dismissal of his appeal against conviction.
7. It is convenient for their Lordships to deal first with the jurisdiction issue.

The jurisdiction issue

8. Their Lordships have no doubt that the Privy Council does have jurisdiction to entertain an appeal by Mr Barlow against the 1996 dismissal by the Court of Appeal of his appeal against conviction. Section 42 of the Supreme Court Act 2003 put an end to appeals to the Privy Council "from or in respect of any … criminal decision of a New Zealand court made after 31 December 2003…" and section 49 provided that, as from 1 January 2004, the Imperial enactments under which appeals from New Zealand courts to the Privy Council had been brought should "cease to have effect as part of the law of New Zealand". But section 52(1)(b) said that applications to the Privy Council for leave to appeal against a decision of a New Zealand court made before 1 January 2004 "must be determined as if sections 42 and 49 had not been enacted". So the jurisdiction of the Privy Council to deal with Mr Barlow's application for leave to appeal against the 1996 Court of Appeal decision was preserved.
9. It is, of course, correct that an examination by the Judicial Committee of the question whether, in the light of the new evidence on which he relies, Mr Barlow's appeal against his conviction should be allowed ultimately raises very much the same question as must have been addressed by the Ministry officials when preparing their report to the Governor-General and by Dr Fisher QC when reviewing that report. Their Lordships would also accept that it is substantially the same question as must have been addressed by the Governor-General in deciding how to deal with Mr Barlow's application to him. But these coincidences of subject matter do not, in their Lordships' opinion, go to jurisdiction. While they may be relevant to how the Board should exercise the jurisdiction preserved for it by the 2003 Act, they cannot negate that jurisdiction. Moreover, if their Lordships were to conclude that the new evidence did indeed show that there had been a miscarriage of justice, neither the opinion of Dr Fisher, nor the report of the Ministry, nor even the Governor-General's decision based on that opinion and report could justify the Board withholding the remedy for that miscarriage to which Mr Barlow would otherwise be entitled.
10. The Crown has also argued that, since it is not suggested that, on the evidence as it stood when the Court of Appeal dismissed Mr Barlow's appeal against conviction in 1996, any error can be discerned in the Court of Appeal's dismissal of his appeal, "no justiciable issue arises engaging the powers of the senior appellate court in either jurisdiction", and, therefore, that "no further appeal can lie" (para 73 of the Crown Case as to Jurisdiction). Walker v The Queen [1994] 2 AC 36, an appeal to the Privy Council from Jamaica in which the Board dismissed the appeal for want of jurisdiction, is relied on in support of this proposition. Walker was a case in which the appeal related to the imposition of the mandatory death penalty. The appellants had been sentenced to death for murder. They had appealed against conviction but their appeals had been dismissed. They had not appealed against sentence, and they could not have done so since the death sentence was mandatory. Their appeal to the Privy Council was based on the long delay to which they had been subjected while awaiting execution, a delay which, they contended, would render their execution unconstitutional. In these circumstances, the appeals being neither appeals against conviction nor appeals against sentence, the Board held that, as an appellate tribunal, it lacked jurisdiction to deal with what was a new issue, namely, the constitutionality of their execution after the long delay. The reasoning in Walker is of no assistance in the present case. The new evidence relied on by Mr Barlow undermines, it is submitted, prosecution evidence that had been placed before the jury and, in the circumstances, his conviction should be quashed. Mr Barlow directly challenges his conviction and the dismissal by the Court of Appeal of his appeal against conviction.
11. Fresh evidence, which is said to show that there has been a miscarriage of justice and which emerges after the trial and an unsuccessful first appeal, can be the basis of an appeal to the Board. None of the authorities referred to by Mr Pike, counsel for the Crown before their Lordships, is inconsistent with that proposition. Neither Thomas v R [1980] AC 125 nor Walker v R [1994] 2 AC 36 was a case in which the appeal sought to be brought before the Privy Council was an appeal against the dismissal by the Court of Appeal of an appeal against conviction or against sentence. Moreover, cases such as Pitman v The State [2008] UKPC 16 show quite clearly that the Board will entertain an appeal on the basis of fresh medical evidence (introduced for the first time before the Board) as to the accused's mental state at the time of the offence. In that case the Board remitted the evidence for the consideration of the Court of Appeal. Plainly, the jurisdiction to entertain an appeal on such a basis is one to be exercised with caution, but that the Board has such a jurisdiction their Lordships have no doubt.
12. In British Coal Corporation v R [1935] AC 500, another case on which the Crown relied, the issue was whether a provision of a Canadian statute purporting to restrict appeals to the Privy Council in criminal matters was within the legislative competence of the Canadian legislature. The Privy Council held that it was. There is no comparable New Zealand statute barring an appeal from the 1996 decision of the Court of Appeal. On the contrary, the jurisdiction of the Privy Council to hear appeals "from or in respect of any … criminal decision of a New Zealand court made" before 1 January 2004 was expressly preserved by the 2003 Act. The most that can be said is that an application under section 406 of the Crimes Act 1961 provides a possible alternative avenue for relief. But that section does not purport to replace the jurisdiction of the Privy Council to entertain an appeal based on fresh evidence emerging after an unsuccessful appeal to the Court of Appeal and their Lordships are not disposed to infer that that was the intention of the section.
13. While the Board will always be astute to ensure that an application to introduce fresh evidence is not used as a vehicle to undermine the finality of trials, no such issue arises in this case. Indeed, the Crown accepts that the evidence was not available at the time of the trial. Their Lordships accordingly consider that they have jurisdiction to entertain Mr Barlow's application for special leave to appeal and that they should grant the application and proceed to consider the merits and the implications of the fresh evidence for his conviction. They must first consider the test which they are to apply in considering these matters.

Section 385(1) of the Crimes Act 1961

14. Section 385(1) provides, so far as relevant for present purposes, as follows :

"(1) On any appeal [against conviction] the Court of Appeal or the Supreme Court must allow the appeal if it is of opinion –

(a) ……..

(b) ……...

(c) That on any ground there was a miscarriage of justice; or

(d) ……..

and in any other case shall dismiss the appeal :

Provided that the Court of Appeal or the Supreme Court may, notwithstanding that it is of opinion that the point raised in the appeal might be decided in favour of the appellant, dismiss the appeal if it considers that no substantial miscarriage of justice has actually occurred."

15. In Bain v The Queen [2007] UKPC 33, on the basis of the submissions made by both parties, the Board applied the well settled interpretation of this section in cases such as R v McI [1998] 1 NZLR 696. And, at the hearing of the present appeal, neither side made any detailed submissions on its interpretation.
16. After the conclusion of the oral hearing, however, Mr Barlow's New Zealand solicitors very properly provided their Lordships with a copy of the subsequent judgment of the Supreme Court of New Zealand in Matenga v The Queen [2009] NZSC 18. In that case, in a judgment given by Blanchard J, the Supreme Court set aside the conviction of the appellant on a charge of rape and directed a new trial. The court placed an embargo on the publication of the reasons for their decision pending a retrial. Their Lordships will accordingly make no reference to the facts. Blanchard J's judgment contains a fresh analysis of section 385(1)(c) and of the way that the proviso in the subsection should be applied.
17. In particular, Blanchard J was concerned to analyse what would constitute "a miscarriage of justice" for the purposes of paragraph (c) and what would constitute a "substantial" miscarriage that had "actually occurred" for the purposes of the proviso.
18. After considering New Zealand case law on section 385, in particular, R v McI [1998] 1 NZLR 696 (CA) and R v Sungsuwan [2006] 1 NZLR 730 (SC), and the judgment of six members of the High Court of Australia in Weiss v R (2005) 224 CLR 300 on the application of a statutory provision in much the same terms as section 385(1), at para 27, Blanchard J said:

"We are persuaded of the soundness of the general approach taken by the High Court in Weiss, which we consider should now be followed in New Zealand, subject to two qualifications which are identified below."

19. The approach of the Australian High Court was that a court ought not to attempt to predict what a jury might or might not have done if the irregularity said to constitute the miscarriage had not happened. Instead the appellate court should itself decide whether a substantial miscarriage of justice had actually occurred. The task was an objective task to be performed on the record of the trial: it was not an exercise in speculation or prediction (see para 24 of Blanchard J's judgment). In para 28 of his judgment Blanchard J summed up the required approach in this way:

"… the decision to confirm a jury verdict, despite something having gone wrong, depends on whether the appellate court considers a guilty verdict was inevitable on the basis of the whole of the admissible evidence (including any new evidence). The Court must also be satisfied that overall there has been a fair trial. The Bill of Rights Act guarantees of a trial by jury and an appeal do not require that a further jury trial should necessarily be ordered if a miscarriage at the first trial has been identified. Nothing in that Act prevents the appellate court from considering whether, despite the miscarriage, the verdict already rendered by a jury should stand."

His Honour expanded on the Weiss approach in paragraphs 30 and 31:

"30. The Weiss Court accepted that a miscarriage under our para (c) is anything which is a departure from applicable rules of evidence or procedure. We have hesitated about whether in its statutory context that is the meaning which should be given to the word, lest it lead to the application of the proviso in a large number of cases. Few trials are perfect in all respects … In the end, departing in this respect from Weiss, we consider that in the first place the appeal court should put to one side and disregard those irregularities which plainly could not, either singly or collectively, have affected the result of the trial and therefore cannot properly be called miscarriages. A miscarriage is more than an inconsequential or immaterial mistake or irregularity.

31. Proceeding in this way and having identified a true miscarriage, that is, something which has gone wrong and which was capable of affecting the result of the trial, the task of the Court of Appeal under the proviso is then to consider whether that potentially adverse effect on the result may actually, that is, in reality, have occurred. The court may exercise its discretion to dismiss the appeal only if, having reviewed all the admissible evidence, it considers that, notwithstanding there has been a miscarriage, the guilty verdict was inevitable, in the sense of being the only reasonable possible verdict, on that evidence. Importantly, the Court should not apply the proviso simply because it considers there was enough evidence to enable a reasonable jury to convict. In order to come to the view that the verdict of guilty was inevitable the Court must itself feel sure of the guilt of the accused. Before applying the proviso the Court must also be satisfied that the trial was fair and thus there was no breach of the right guaranteed to the accused by s 25(a) of the Bill of Rights".

20. In para 33 of the judgment, Blanchard J explained the second respect in which the Weiss approach needed to be qualified:

"The High Court said that the appellate court's task under the proviso was to be undertaken on the whole of the record. That is correct. However, it expressly included in the record the fact that the jury has returned a guilty verdict. Whilst the verdict may indicate the jury's view on some question unrelated to the miscarriage, the appeal court must form its own view on whether a finding of guilt was, notwithstanding the miscarriage, the only reasonably possible verdict."

21. Not having been addressed by counsel on the point, their Lordships express no view as to whether and, if so, to what extent, there is a conflict between the approach prescribed by the Supreme Court in Matenga and the approach adopted by the Board in Bain. But they are satisfied that, even if the approach in Matenga had been applied, having regard to the new evidence in that case, there would have been no room for the application of the proviso and the result of the appeal would have been the same. However that may be, in the absence of any submissions to the contrary, and without reconsidering the point for itself, the Board considers it appropriate to apply the approach of the Supreme Court in Matenga to the interpretation and application of section 385(1)(c) and the proviso.
22. Following the guidance given in Matenga, the first step is to identify whether something went wrong at the trial which was "capable of affecting the result of the trial", and then to consider under the proviso "whether that potentially adverse effect on the result may actually, that is, in reality, have occurred." With these steps in mind, their Lordships now turn to examine the evidence at the trial and the significance of the new evidence on which Mr Barlow relies.

The evidence at the trial

23. Mr Barlow had a business relationship with Mr Eugene Thomas and Mr Gene Thomas. The details of the relationship are not clear from the evidence save that it appears to be the case that he owed them $70,000. He was, however, "financially comfortable" (para 43 of the Petition) and no possible reason why he should have wanted to murder them has ever been suggested.
24. The Thomases had an office in a building, Invincible House, in Wellington. The office was not on the ground floor and could be reached either by stairs or by a lift. A meeting between Mr Barlow and the Thomases in their office was specially arranged for 5.30pm in the evening of Wednesday 16 February 1994. The evidence does not disclose the purpose of the meeting. It must, of course, be known to Mr Barlow. The meeting was entered in five diaries and was known to a number of people.
25. Mr Thomas snr had been gardening that afternoon and so he changed out of his gardening clothes and travelled into town for the meeting. Mr Thomas jnr, who had been to the dentist, also made his way to their office in Invincible House. At about 5.40 pm Mr Thomas jnr telephoned his brother to say that he would be running late and would not be able to collect recording equipment needed for a meeting at 7.30 pm. According to his brother, it would have been uncharacteristic of Mr Thomas jnr to schedule appointments he could not keep and so he thought that something unexpected must have arisen to delay his brother.
26. At about 6.40 pm Mr Barlow was seen by office cleaners leaving the ground floor entrance to Invincible House. He was carrying a briefcase. Shortly after that, the bodies of the Thomases were found by the cleaners in or near their boardroom. Both had been shot.
27. It appeared that Mr Thomas jnr, who had first been shot in the hand, had managed to retreat to a reception area outside the boardroom where he had been killed by a shot to the head. Mr Thomas snr had been shot from the front while sitting at the boardroom table and had then been shot from behind through the back of the neck – apparently while attempting to make a telephone call. Someone had torn the top pages from a diary of one or other of the Thomases in which the appointment with Mr Barlow had been noted. The fact of the appointment was, however, apparent from indentations discernible on the uppermost of the remaining pages of the diary. Pages from a notepad that had been in front of Mr Thomas jnr had also been removed. Two fingerprints made by Mr Barlow were found on the notepad and there was evidence, albeit disputed, that the position of the prints was consistent with his having torn off a page or pages from the pad.
28. The police interviewed Mr Barlow. He gave two contradictory explanatory statements to them.
29. In his first statement, made on 17 February, the day after the murders, Mr Barlow said he had arrived at the Thomases' offices for the meeting and had been asked by Mr Thomas jnr to wait. He said that he heard, through a closed door, Mr Thomas snr talking to someone. Mr Barlow said that, after waiting some time, he had left but, having driven some way towards his home, had changed his mind and returned to the offices. On his return he said that he had not seen either of the Thomases. Mr Thomas snr had called out to him that he (Mr Thomas) would not be able to see him that day. He said he had then left the premises for a second time, passing some office cleaners as he did so and that the time of his second departure was about 6.30 pm.
30. In his second statement, made a day later when he was re-interviewed by the police, Mr Barlow repeated his first story up to the point when he had re-entered Invincible House and gone up to the Thomases' office for a second time. He now said that he had come upon the body of Mr Thomas jnr in the reception area. He had stepped over the body and gone into the boardroom where he had found the body of Mr Thomas snr. He said he had not touched either body but had left the premises via the stairs, passing the office cleaners on the way. He had said nothing to them. He did not raise the alarm or seek medical assistance for the Thomases.
31. At an early stage in their investigations police officers searched Mr Barlow's car. In it they found a receipt which led them to conduct a search of a local refuse tip, the Happy Valley tip. The search, that lasted through the weekend, uncovered a CZ 27 pistol, fitted with a homemade (22 barrel in place of its original (32 calibre barrel, the remains of a homemade silencer and a box of Geco brand (32 calibre lead-nose ammunition. The box would originally have contained 50 bullets, but there were 43 in it when it was found. It is accepted that Mr Barlow threw all these items into the refuse tip the morning after the murders.
32. It is worth recording, en passant, that there was evidence to the effect that only 43 boxes (each containing 50 bullets) of Geco ammunition had ever been imported into New Zealand and 6 of those 43 may not have contained lead-nose bullets.
33. Not only had the original barrel of the CZ 27 pistol been removed, but the original magazine had also been removed and a homemade magazine substituted. In addition, the firing pin of the pistol had been filed down as, also, had the cartridge case extractor claws. The prosecution case was that Mr Barlow had made these alterations in order to prevent the identification of the CZ 27 pistol as the murder weapon.
34. Mr Barlow, who did not give evidence at the trial, did not explain to the police why he had thrown away his CZ 27 pistol, the silencer and the Geco (32 ammunition. Nor did he tell them what he had done with the pistol's original (32 barrel. He did, however, give some explanation of these matters in the course of telephone conversations with a business associate and friend of his, a Mr B. Unbeknownst to Mr Barlow a number of these conversations were recorded by the police. The recordings were played to the jury and Mr B gave evidence for the prosecution.
35. In the course of these conversations Mr Barlow told Mr B that, when he had re-entered the offices and come across the bodies of the two Mr Thomases, he saw his CZ 27 lying on the floor with the silencer attached. He had remembered that, some time previously, he had lent the pistol to Mr Thomas snr, who had been receiving threats from someone or other, for his protection. He told Mr B that he had been concerned that his fingerprints might be on the pistol or that it might otherwise be traced to him and that he would blamed for the shooting. So, he said, he had retrieved the pistol and silencer and had thrown them away, with the box of Geco bullets, in the Happy Valley tip. Before doing so, he had substituted a (22 barrel for the original (32 barrel, had filed down the pistol's firing pin and had cut up the rubber lining of the silencer. Mr Barlow suggested, without positively so stating, that he had disposed of the original (32 barrel somewhere in the Wellington Botanical Gardens. Despite a subsequent search of the Botanical Gardens, the police failed to find the barrel.
36. It is right to record that at no time in the course of his recorded conversations with Mr B did Mr Barlow say anything constituting an admission that he had killed the Thomases. And, despite an extensive police examination of Mr Barlow's clothing and effects, no traces of blood could be found, save on the briefcase that Mr Barlow had taken with him to the intended meeting. That blood could not be positively identified as coming from either of the Thomases.
37. While Mr Barlow stated in his conversations with Mr B that he had found the CZ 27 pistol on the floor, the defence contention was that it had not been used to carry out the killings. Rather, the two men had been killed by a third person unknown, who had used another weapon, probably a revolver, that had fired the bullets found at the scene. The defence theory was that, in some kind of an attempt to defend himself, Mr Thomas snr had taken out the pistol which, Mr Barlow said, he had lent to him some months previously and which he had found on the floor.
38. It is clear to the Board, as no doubt it was clear in each of the three trials and to the Court of Appeal, that if Mr Barlow was to be convicted, the jury would have to be satisfied beyond reasonable doubt on the evidence as a whole, that, contrary to the submissions made on behalf of Mr Barlow, the CZ 27 pistol was indeed the murder weapon. More particularly, the case against Mr Barlow would collapse if it could be shown that the pistol could not have been the murder weapon. In these circumstances the forensic science evidence relating to the murder weapon and the bullets was plainly of importance.
39. Each of the four scene bullets had been recovered and examined by forensic scientists. They were (32 calibre bullets. The prosecution needed to satisfy the jury that the bullets that killed the two Mr Thomases had been, or could have been, fired from the CZ 27.
40. An examination of the original barrel of the CZ 27 and its rifling characteristics would almost certainly have allowed an expert to give a confident opinion as to whether the bullets recovered at the murder scene had passed through the barrel. But no examination of that kind was possible because Mr Barlow had removed and concealed the barrel.
41. There was, however, firm evidence that the murder weapon had been fitted with a silencer. If not, the four shots would have been very audible - and no one in the vicinity at the time of the murders had heard any shots. Moreover, Mr Wilson, a forensic scientist who gave evidence for the Crown, testified that the nature of the wound to Mr Thomas snr's temple indicated that a silencer had been in contact with the temple when the shot was fired (p 456, 15/20). Therefore, if the CZ 27 was the murder weapon, the bullets must also have passed through the silencer fitted to it. Mr Barlow had said that he had recovered the silencer fitted to the CZ 27 at the scene and the remains of a home-made silencer were found at the tip. So forensic science evidence relating to the question whether the scene bullets had passed through this silencer was potentially important.
42. The evidence for the prosecution on that matter was given by Mr Wilson. He referred to traces of rubber that he had found in the wounds of the two victims and on some of the scene bullets. There was an issue as to whether the composition of these traces of rubber was consistent with that of the rubber lining of the silencer. Mr Wilson thought that it was but there was a clash of expert evidence about this which their Lordships are unable to resolve for themselves. In other words, while they cannot say positively that the composition of the traces of rubber on the scene bullets is consistent with the composition of the rubber lining of the silencer, equally, they cannot say that it is inconsistent with the composition of the rubber lining. This is a factor which the Board will have to take into account when it comes to consider the evidence as a whole.
43. Evidence about the composition of the scene bullets themselves was important. If they could be shown to be similar in composition to the bullets in the Geco box disposed of by Mr Barlow in the Happy Valley tip, the prosecution case that the scene bullets came from that box would be strengthened. Conversely, if the composition of the scene bullets were shown to be sufficiently different from the composition of the bullets in the Geco box to make it unlikely that they came from that box, the prosecution case would be weakened.
44. The new evidence relied on to challenge Mr Barlow's conviction relates to the question whether the scene bullets were, or could be, Geco bullets.
45. At the third trial the Crown led evidence about the composition of the four scene bullets from a Mr Charles Peters, who worked as an analyst for the FBI. He explained that, once manufactured, bullets would be put in storage by the manufacturer before being boxed. Consequently, a box of bullets might contain bullets from batches of bullets manufactured at different times and therefore varying, to some extent, in their composition. He said "we expect several compositions within one box of ammunition" (p 558, 17). As to Geco ammunition in general, he said "we find between 3-4 compositions in a box of Geco" (p 558, 32/33). As to the 43 bullets in the Geco box found at the tip, he said he had analysed a sample of them and had found two compositions (p 558, 31). The four scene bullets were numbered 2030, 2031, 1026, 3060. The first two were those that struck Mr Thomas snr. The other two were those that struck Mr Thomas jnr. Bullets 2030, 1026 and 3060 were, he said, similar in composition to one another and similar to 14 of the sampled bullets in the discarded Geco box. He was asked whether these three scene bullets "could have come from the box" found at the tip. His answer was "Yes" (p 589, 8/9). Bullet 2031 had, he said, a composition which was not the same as the other scene bullets and which did not match either of the two compositions found in the sampled group of the box bullets. He was, of course, asked about the significance of this. He responded -

"The only significance to 2031 is that as I explained … in ammunition there is more than one composition. This [2031] certainly could have come from the same box of ammunition as the other samples. It is just that we cannot positively say, as we have nothing to match it with. These are the types of differences we see within a box ….. Maybe if we analysed the remaining cartridges in this box this composition may have come up … (p 559, 28/35).

46. Mr Peters also gave evidence that the FBI maintained a database of the various compositions of bullets that had been analysed by the FBI. These included Geco bullets. Mr Peters was asked whether the composition of the scene bullets matched any composition of bullets recorded on the FBI database. He said:

"We did find this composition to come out in our database and it came up with some other (32 automatic auto Geco ammunition and nothing else."

The questioning continued:

"Q. Nothing other than Geco ammunition?

A. Yes.

Q. Is there anything your database does not include?

A. I am sure there is but it includes almost everything" (p 559, 24/35)

47. The evidence of Mr Peters to which reference has been made was given in the course of his examination-in-chief. At the conclusion of his examination-in-chief he was asked to summarise his findings. He responded:

"Basically the bullet lead we analysed could be divided into three compositional groups, A [i.e. 2031], B [i.e. 2030, 1026, 3060 and 14 of the box bullets] and C [i.e. the rest of the box bullets]. The specimens in group B and the specimens in group C, within their individual group, are analytically indistinguishable from one another, in other words they match in elemental composition. Furthermore, it is my opinion that they came from the same manufacturer and are of the same type and the same source of lead at that manufacturer."

He was then asked whether they could have come from the same box and he replied:

"They came from this, the same box or a box manufactured on or, manufactured or loaded on or about, that same date."

In the second sentence of the first of these quotations Mr Peters had stated that the composition of scene bullets 2030, 1026 and 3060 was "analytically indistinguishable" from the composition of 14 of the Geco bullets found at the tip. In the last sentence of that quotation he expressed the opinion, consequential upon his compositional comparison, that all these bullets came from the same manufacturer, i e that the three scene bullets, like the 14 tip bullets, were Geco bullets. In his further answer, he said that the three scene bullets came from the same box, or from a box manufactured on, or loaded on or about, the same date.

48. The evidence about the bullets and the silencer was, from the prosecution's point of view, necessary for the purpose of refuting any suggestion that Mr Barlow's CZ 27 pistol could not have been the murder weapon. The prosecution needed to show that the pistol could have been the murder weapon and that the scene bullets could have been Geco bullets taken from the box before it was thrown in the tip. In other words, if the Crown failed to surmount these hurdles, its case must fail.
49. But, as the trial judge explained in opening the part of his summing up devoted to the scientific evidence relating to the bullets, the Crown actually sought to derive positive assistance from that evidence:

"The Crown has sought to show that analysis of the materials of the bullets from the scene and from the box at the tip showed such a relationship in their content that you could conclude it was likely the bullets came from the same box, i e that the scene bullets, likely that the scene bullets had been taken from Mr Barlow's box if you accept it was indeed his box in the tip."

50. So, on the Crown approach, the comparative analysis of the composition of the three scene bullets and of the bullets in the box in the tip not only showed that the CZ 27 pistol could have been the murder weapon, but was actually a positive factor pointing to the conclusion that it was the murder weapon. Of course, the scientific evidence was simply one item in a much larger body of circumstantial evidence, which the Board will have to mention in due course, on which the Crown relied in order to establish that Mr Barlow had used the pistol to kill the Thomases.

The New Evidence

51. Mr Peters' evidence about the scene bullets and the Geco bullets found at the tip was based on the technique of comparative bullet lead analysis ("CBLA"). He had given similar evidence based on that technique in a number of trials in the United States, some of which had resulted in convictions. As Mr Pike very freely accepted on behalf of the Crown, recently conducted studies and research in the United States has undermined the usefulness of this technique for the purpose of identifying the source or type of bullets used in a crime. In consequence, the FBI had ceased to use the techniques for that purpose. In the printed case for the appellant Mr King referred to State of New Jersey v Behn 7 March 2005, in which the Appellate Division of the Superior Court of New Jersey set aside a conviction after a trial in which Mr Peters had given CBLA evidence. The Court commented that

"… the integrity of the criminal justice system is ill-served by allowing a conviction based on evidence of this quality whether described as false, unproven or unreliable to stand."

52. In Behn and, apparently, in a 2003 appeal in USA v Mikos, both of which cases involved challenges to the testimony of Mr Peters given at trial, the key defence expert was a Dr Randich. Dr Randich has made a report dated 26 March 2005, sworn as an affidavit on 15 January 2009, in support of Mr Barlow's application and appeal in the present case. The affidavit includes this conclusion :

"I want to make to make it clear that on the basis of the analysis undertaken by Mr Charles A. Peters it is my firm opinion that it is simply not possible to say who manufactured the scene bullets. In my opinion it cannot be said that Geco produced the scene bullets. There is no reason to believe, from a compositional analysis standpoint only, that the bullets and/or lead fragments recovered from the bodies or crime scene in the Barlow case are Geco brand bullets and no other brand. The compositions of the lead fragments found at the scene are consistent with the Geco ammunition under control of the suspect but that is all one can say about it."

53. On the basis of Dr Randich's report, Mr Pike accepted that Mr Peters went too far when he said "Furthermore it is my opinion that they [i e bullets 2030, 1026, 3060] came from the same manufacturer and are of the same type and the same source of lead at that manufacturer" and when he added that they came from the same box as the one in the tip or from a box manufactured or loaded on or about the same date.
54. On the other hand, Dr Randich's report does not undermine Mr Peters' evidence that the three scene bullets were analytically indistinguishable from those in the box which Mr Barlow threw away in the Happy Valley tip. On the contrary, Dr Randich confirms that the lead fragments found at the scene were consistent with the Geco ammunition which Mr Barlow had. Dr Randich was objecting to the further proposition that that consistency justified the conclusion that the scene bullets were Geco bullets.
55. In short, Mr Peters' evidence - that the three scene bullets were Geco bullets and, a fortiori, that they came from the box found in the tip or were manufactured or loaded about the same time – is shown to be unscientific and untenable. But his evidence that the three scene bullets were consistent in composition with Geco bullets is not undermined: rather, it is supported by Dr Randich.

Was there a miscarriage in terms of section 385(1)(c)?

56. It follows that what went wrong at the trial was the leading of Mr Peters' evidence that the three scene bullets were Geco bullets and, a fortiori, that they came from the box found on the tip or were manufactured or loaded about the same time. No such evidence should have been led. The only evidence which should have been led from Mr Peters was his evidence that the three scene bullets were consistent in composition with Geco bullets. Therefore, this appeal, although based on fresh evidence, is similar to an appeal on the basis that some of the evidence led on behalf of the Crown was inadmissible in law.
57. Returning to the test in Matenga v The Queen [2009] NZSC 18, the Board must therefore first consider whether the leading of this evidence from Mr Peters was capable of affecting the result of the trial. Their Lordships are satisfied that it was, because that element in Mr Peters' evidence contributed materially to the body of scientific evidence which, the Crown said, permitted the jury to conclude that "it was likely that the bullets came from the same box" as the one found in the tip. As already explained, if the jury accepted the scientific evidence that the three scene bullets did actually come from the box on the tip, that could provide an independent element in the body of circumstantial evidence against Mr Barlow. Removing a material element from that body of circumstantial evidence must in principle be capable of affecting the result of the trial.

The Proviso

58. The Board must therefore proceed to the next stage and consider the proviso. Of course, as the Supreme Court pointed out in Matenga, at para 31 (quoted at para 19 above) there can be no question of the Board applying the proviso if what went wrong at the trial made the trial unfair. So the first thing to be considered is whether the defect in the trial was such as to render it unfair. In his able submissions on behalf of the appellant, Mr King argued that the introduction of this part of Mr Peters' evidence did indeed have that effect. He drew attention to the passage in the judgment of the Board in Bain v The Queen [2007] UKPC 33 where Lord Bingham of Cornhill said, at para 115:

"…a fair trial ordinarily requires that the jury hears the evidence it ought to hear before returning its verdict, and should not act on evidence which is, or may be, false or misleading. Even a guilty defendant is entitled to such a trial."

In effect, Mr King's submission was that in this case the jury had been invited to act on evidence given by Mr Peters, a material part of which, as the Crown now acknowledges, was misleading. While the Board accepts that this was indeed what happened, it is certainly not the case that a trial is rendered unfair simply because some potentially misleading evidence has been admitted. The fairness of the trial has to be judged in the light of the proceedings as a whole. In this case, in their Lordships' view, the introduction of the evidence was not "such a departure from the essential requirements of the law that it [went] to the root of the proceedings": Wilde v The Queen (1988) 164 CLR 365, 372, per Brennan, Dawson and Toohey JJ, approved by the Board in Howse v The Queen [2006] 1 NZLR 433 (PC).

59. Under the proviso, therefore, the Board must now consider whether the potentially adverse effect of the introduction of this part of Mr Peters' evidence on the result of the trial "may actually, that is, in reality, have occurred." In doing so, the Board must consider whether, in the light of all the admissible evidence, notwithstanding the miscarriage of admitting the unsustainable part of Mr Peters' evidence, the guilty verdict was "inevitable, in the sense of being the only reasonably possible verdict, on that evidence."
60. For this purpose the Board has to take account of the relevant evidence as a whole, both that tending to incriminate Mr Barlow and that in his favour. A convenient summary of the main points on which the Crown relied at the trial can be found in Neazor J's summing-up, at pp 39-40 of the transcript.

"The Crown case … relies on 12 points of significance. Firstly, this was an after hours meeting set up for Mr Barlow; secondly, the accused was seen leaving minutes before the bodies were found; thirdly, he altered the CZ pistol and silencer and disposed of them with ammunition within a few hours; he disposed of the (32 barrel separately. Next, the scene reconstruction suggests there was a meeting in progress in the boardroom when the shootings occurred and that the removal of the diary entry for Mr Barlow was significant, as were his prints on the pad, that is, the suggested reason for them being there. Next, Mr Barlow had a long association with the victims, and it is suggested that he had some matter that he was going to discuss with them. Next, he made two inconsistent written statements to the police, followed by a third different statement to Mr B in August; that what he did, his actions were an elaborate scheme of deception and that included going to the tip and dumping the gun, the silencer and the ammunition. Next, that the CZ 27 when it had had a (32 barrel had been fired by him and he acknowledged that to Mr B. Eight, that there were other unlicensed handguns in the house but only the CZ 27 was found in the tip. Ninth, he was involved with the making of the silencer through his family; why dispose of it if it had not been used in the shooting? Ten, scientific evidence about luminol tests and the connections in relation to exhibits that were referred to by Mr Wilson in the chart that he put in. Next, what is on the B tapes which [counsel for the Crown] said shows much of Mr Wilson's work is correct by Mr Barlow's own admission. Twelve, he said: where is the barrel of the CZ 27 which might have been so revealing one way or the other? [Counsel for the Crown] said those are the major pegs of the Crown case and he asked you to come back to them all the time. He said when you take them all together they point beyond reasonable doubt to the accused's guilt."

61. As the Board has pointed out already, an earlier passage in the summing-up shows that the Crown did rely on the forensic science evidence relating to the bullets to suggest that the jury could conclude that it was likely that the scene bullets had been taken from the box in the tip. Nevertheless, it is right to notice that, according to this summary – whose accuracy has not been challenged – counsel for the Crown did not include the comparison evidence relating to the bullet, far less the unsound part of Mr Peters' evidence, among the main points of his case, to which he asked the jury to come back all the time.
62. The summing-up does not contain a similar concise summary of the main points on which defence counsel relied. And, of course, there is no need for a defendant to produce a coherent picture in response: all that he need do is persuade the jury that the evidence for the Crown leaves them with a reasonable doubt as to the defendant's guilt. A number of obvious points which defence counsel made are easily identified, however. First, the Crown had not established any motive which would explain why Mr Barlow would have killed the Thomases. The apparent debt of $70,000 was unconvincing, since there was nothing to suggest that Mr Barlow would not have been in a position to repay it. Secondly, if Mr Barlow had indeed killed the Thomases, he must have set out armed with the pistol and with the intention of doing so. If so, why make an appointment in advance and one which was known to other people? Thirdly, since the appointment was known and was entered in no less than five diaries, what would have been the point, for him, of tearing out pages from the diary and notebook? Fourthly, if, as his brother thought, Mr Thomas jnr's telephone call at 5.40 pm indicated that something unexpected had arisen to delay him, it could not have been the scheduled appointment with Mr Barlow. So something else must have come up. Fifthly, there was nothing by way of physical evidence to associate Mr Barlow with the killing. In particular, there were no bloodstains on him and the blood on his case could not be linked to the victims. The fingerprints on the notebook showed nothing more than that he had been in the room – something which he had in any event admitted in his second statement to the police. Sixthly, the tests done on Mr Thomas snr's hands were consistent with him having handled a firearm on the day he died and this tended to support the defence contention that Mr Thomas had taken out the pistol in some kind of attempt to defend himself. Seventhly, in the various telephone conversations Mr B had induced Mr Barlow to say a great deal about what happened. Quite a lot of what he said was confirmed by the other evidence in the case. But at no stage had Mr Barlow ever admitted killing the Thomases. Eighthly, counsel suggested that the police had concentrated on Mr Barlow from an early stage in their investigation. This meant that they had not explored any other lines which might have revealed who had actually carried out the killings, on which Mr Barlow said he had stumbled when he went to the office. Ninthly, if Mr Barlow had indeed found his pistol on the floor of Mr Thomas snr's office, then, it was understandable, even if unwise, for him to have panicked and, instead of contacting the police, to have removed the pistol and taken the various steps which he did in relation to it. Moreover, he had been quite willing to submit to various tests when asked by the police to do so.
63. In addition, it is plain, both from the extended passage in the judge's summing-up at pp 30-33 of the transcript and from summary of defence counsel's closing submissions at pp 41-42, that defence counsel emphasised the criticisms of the Crown forensic science evidence, including the evidence of Mr Peters, which had been made by the experts, Mr Barnes and Dr Reeves, called for the defence. In particular, defence counsel used certain German tests, which the Crown had relied on at the previous trial, to criticise the comparison evidence of Mr Peters. Counsel for the Crown pointed out, of course, that defence counsel was now relying on the German evidence which, in the previous trial, the defence had criticised as unreliable.
64. Their Lordships note, in particular, the sharp difference of opinion between Mr Wilson, the Crown expert, and Dr Reeves, the defence expert, as to whether the traces of black sealant found on the scene bullets could have come from a Geco bullet – as Mr Wilson said. But they also note the Crown evidence that, with the benefit of enquiries made abroad, the only bullets with similar characteristics to the scene bullets which were known to have a sealant were Geco bullets.
65. Their Lordships have considered the available evidence, minus the unacceptable evidence of Mr Peters. While, as they have just mentioned, there was indeed evidence from the German database which suggested that it could be said, positively, that the three scene bullets were Geco bullets, their Lordships think it preferable, in the circumstances, to ignore that particular item of evidence. Considering the other scientific evidence as a whole, and taking due account of the disputes between the experts for the Crown and the defence, they are satisfied that the evidence shows that the three scene bullets could have been Geco bullets. So these bullets could have come from the box of bullets found in the Happy Valley tip.
66. Having reached that conclusion, their Lordships have gone on to consider the rest of the evidence for the Crown and the criticisms which can be made of that evidence. Despite those criticisms, the Board finds that the circumstantial case against Mr Barlow is overwhelming.
67. While all the points made on behalf of the Crown were significant, their Lordships are particularly struck by certain very obvious aspects of the circumstantial case against Mr Barlow.
68. First, Mr Barlow had been present in the room in which the murders were committed at about the time they were committed. There was no evidence that anyone else was there although Mr Barlow had told the police that he had heard Mr Thomas snr speaking to someone. Nor, of course, was there any evidence identifying a third person who would have had a motive for committing the murders.
69. Secondly, the day after the murders were committed, Mr Barlow went to the refuse tip and threw away his CZ 27 pistol, the silencer that, on his own account, had been attached to it and his box of Geco ammunition, which lacked seven bullets. Since the pistol could have fired the bullets which killed the Thomases, disposing of these items tends to point to his involvement in the murders.
70. Thirdly, before disposing of the pistol, Mr Barlow had substituted another barrel for the original barrel, had filed down the firing pin and had substituted another magazine for the original magazine. The effect of doing so was to diminish the chances of any connection being made between the killings and the CZ 27 pistol, if it were subsequently found. Moreover, by disposing of the barrel in some unknown place, he deliberately concealed an item which, if tested, could almost certainly have settled one of the key issues in the case. Again, while other explanations for those actions are, of course, possible and were suggested by defence counsel at the trial, the most obvious explanation is that Mr Barlow took those steps to try to dispose of potentially incriminating evidence against him.
71. Fourthly, Mr Barlow gave three contradictory and inconsistent accounts of his conduct after arriving at Invincible House for his meeting with the Thomases – two to the police and a third to Mr B.
72. These are merely four very broad aspects of the circumstantial evidence which stand out. But the evidence relating to the three scene bullets has to be seen in the context of all the other circumstantial evidence. Even if the evidence on the three scene bullets is taken to go no further than to show that they could have been Geco bullets, when that evidence is fitted into the entire body of circumstantial evidence, the Board cannot avoid drawing the inference that the three scene bullets were indeed Geco bullets and that Mr Barlow used the pistol to kill the Thomases. He then took very elaborate steps to get rid of the physical evidence which could link his pistol and bullets to the murders. Most significantly, of course, he disposed of the barrel of the pistol in such a way that it has never been found.
73. Their Lordships emphasise that, when considering the evidence, they have, in particular, had due regard to the defence criticisms of certain aspects of the forensic science evidence led for the Crown. They have also borne in mind the fact that the Crown has established no motive for Mr Barlow killing the Thomases. But, when all these factors are weighed in the context of the evidence as a whole, they do not cause their Lordships to have any reasonable doubt about Mr Barlow's guilt. It is, of course, the case that two previous juries failed to reach a verdict and that the jury in the third trial took many hours to reach their guilty verdict. In Matenga, at para 33 (quoted at para 20 above), the Supreme Court held that an appeal court applying the proviso should not take account of the trial jury's guilty verdict. Similarly, their Lordships consider that they should not take account of the fact that the first two juries failed to agree on a verdict.
74. According to the test laid down in Matenga, at para 31 (quoted at para 19 above), before it can apply the proviso, the Board must consider that, although there was a miscarriage, "the guilty verdict was inevitable, in the sense of being the only reasonably possible verdict, on the evidence." The Board has indeed concluded that, on the evidence as a whole, they have no reasonable doubt about Mr Barlow's guilt and that the guilty verdict was indeed the only reasonably possible verdict. Not only was there enough evidence for a reasonable jury to convict, but the Board itself feels sure of Mr Barlow's guilt.

Conclusion

75. The Board accordingly concludes, in terms of section 385(1) of the Crimes Act 1961, that, while the introduction of the misleading evidence of Mr Peters was indeed a miscarriage, no substantial miscarriage of justice actually occurred.
76. For these reasons their Lordships will humbly advise Her Majesty that Mr Barlow's application for special leave to appeal should be granted, but that his appeal should be dismissed. The parties will have 21 days within which to make any submissions about costs.