Roigard v R
[2020] NZSC 94
•14 September 2020
FOR A COURT READY (FEE REQUIRED) VERSION PLEASE FOLLOW THIS LINK
| NOTE: HIGH COURT ORDER PROHIBITING PUBLICATION |
| IN THE SUPREME COURT OF NEW ZEALAND I TE KŌTI MANA NUI |
| SC 25/2019 [2020] NZSC 94 |
| BETWEEN | DAVID NOEL ROIGARD |
| AND | THE QUEEN |
| Hearing: Further submissions: | 3 October 2019 |
Court: | Winkelmann CJ, Glazebrook, O’Regan, Ellen France and Williams JJ |
Counsel: | R M Lithgow QC and N Levy QC for Appellant |
Judgment: | 14 September 2020 |
JUDGMENT OF THE COURT
The appeal is dismissed.
____________________________________________________________________
REASONS
| Para No. | |
| Glazebrook, O’Regan and Ellen France JJ | [1] |
| Winkelmann CJ and Williams J | [80] |
GLAZEBROOK, O’REGAN AND ELLEN FRANCE JJ
(Given by Ellen France J)
Table of Contents
| Para No. | |
| Introduction | [1] |
| Background | [4] |
| Narrative of events | [5] |
| The Crown case against the appellant at trial | [15] |
| The inmate evidence | [26] |
| The defence case at trial | [36] |
| The Court of Appeal decision | [38] |
| Exclusion under s 8 of the Evidence Act | [40] |
| Context | [40] |
| The submissions | [46] |
| Principles | [53] |
| Assessment | [57] |
| The application of s 30 of the Evidence Act | [72] |
| Result | [79] |
Introduction
The appellant, David Roigard, was convicted after trial of the murder of his son, Aaron Roigard.[1] The appellant was also convicted of eight charges of theft in a special relationship. He was sentenced by the trial Judge, Heath J, to life imprisonment with a minimum period of imprisonment of 19 years.[2] Mr Roigard’s appeal against conviction and sentence was dismissed by the Court of Appeal.[3] He was granted leave to appeal to this Court against conviction.[4]
[1]We refer to David Roigard either as Mr Roigard or the appellant and to his son as Aaron.
[2]R v Roigard [2016] NZHC 166.
[3]Roigard v R [2019] NZCA 8 (French, Cooper and Clifford JJ) [CA judgment].
[4]Roigard v R [2019] NZSC 63 [Leave judgment]. Leave to appeal against other matters including sentence was declined: at [4]–[8].
Aaron’s body has not been found. The case at trial against Mr Roigard was largely circumstantial. This appeal focuses on the evidence called by the Crown from two men who had been in prison with Mr Roigard. The two men are referred to as Mr F and Mr W, as both have name suppression. They gave evidence as to admissions they said Mr Roigard had made to them. The appeal raises the question of whether the evidence of these admissions was properly admitted.[5] This requires us to consider issues about the approach to the exclusion of evidence from such witnesses which are similar to those being considered on the appeal in W (SC 38/2019) v R. Judgment in that case is being delivered contemporaneously.[6]
[5]The approved question is “whether the Court of Appeal erred in upholding the admissibility of the … evidence of the witnesses F and W”.
[6]W (SC 38/2019) v R [2020] NZSC 93 (details of which are suppressed until final disposition of trial). It proved not possible to hear the two cases together but counsel in this appeal were provided with a transcript of the earlier hearing in W (SC 38/2019) v R.
In particular, the appeal will turn primarily on whether the evidence of the two prison informants should have been excluded under s 8(1)(a) of the Evidence Act 2006 on the basis that its probative value was outweighed by the risk that it would have an unfairly prejudicial effect on the proceeding. There is also an issue as to whether the evidence should have been excluded under s 30 of the Evidence Act on the ground that it was improperly obtained. Both issues arise in part from the appellant’s submissions that the system under which witnesses like Mr F and Mr W receive a benefit for giving their evidence, by way of a reduction in sentence, affects the reliability of their evidence and operates unfairly. We address each issue in turn after setting out the background.
Background
The relevant background facts are set out in the judgment of the Court of Appeal and that description is largely adopted in the summary which follows.[7]
Narrative of events
[7]CA judgment, above n 3, at [10]–[24].
Aaron was 27 years of age at the time he disappeared. He lived with his partner, Julie Thoms, in rural Taranaki. The couple had two sons. Aaron worked on a dairy farm for a number of years. He was described as a person with “moderate intellectual limitation” and because of that Mr Roigard helped him with financial matters.[8] Since at least 2007, Aaron had been paying a “significant proportion” of his earnings into what he understood would be an investment that would assist him in realising his dream of owning his own farm.[9] The payments were made to Mr Roigard’s personal account under the description “Sovereign”. It was Aaron’s understanding that Mr Roigard had invested the money on his behalf. Ms Thoms understood that Mr Roigard and Judith Armstrong were looking after the investment for Aaron. (Mrs Armstrong, and her husband Ian, were Mr Roigard’s employers. Aaron had been employed by sharemilkers to assist them on the Armstrong farms and on one occasion he had done some work directly for the Armstrongs.) In the period from January 2007 to April 2014, over $66,000 was paid into the account.
[8]At [10]–[11].
[9]At [11].
Aaron made his last payment to the “Sovereign” account on 2 April 2014. Ms Thoms said it was at about that time that they understood the money would become available to them to buy their own farm property. Both resigned from their jobs in November 2013 anticipating that fact, although both were unaware at that point of an available farm.
There was in fact no accumulated investment. Mr Roigard spent the money on himself. He gave various explanations as to why the investment was not available in April 2014. He got in touch with a real estate agent about available properties and told Aaron about a farm property that would be purchased with the investment money. Later, he told Aaron that the purchase had fallen through and showed Aaron a farm he said would be purchased instead.
On the morning of Monday 2 June 2014 Aaron left his home and drove to his parents’ house.[10] He thought that when he got there, he would be meeting the Armstrongs and signing papers to confirm the purchase of a farm. He and Ms Thoms had started to pack up their house and he had changed his Facebook status to that of “Farm owner”. Aaron made arrangements with his friend, Clinton Bevans, to be available from 10 am that morning to assist with the move. Mr Bevans drove to Aaron’s home but when he arrived Ms Thoms told him that things were running behind schedule and that Aaron was with Mr Roigard. At 11.12 am, Ms Thoms received a text from Aaron’s cellphone which said “They here”. At 12.02 pm, Aaron telephoned Ms Thoms on her cellphone about arrangements for a horse float to assist with the move. She gave the phone to Mr Bevans and Aaron said that the Armstrongs had still not arrived. Mr Bevans could hear Mr Roigard in the background saying “We’ll give them half an hour”.
[10]We refer to this property as the Roigards’ house or the Roigard farm although in fact it belonged to the Armstrongs. Mr and Mrs Armstrong employed Mr Roigard as a maintenance manager of their farm properties and he was provided with accommodation.
Mr Bevans’ cellphone received a text message from Aaron’s cellphone at 12.53 pm which said, “Ther here now f…king tme old man giving sht 2 thm”. In fact, the Armstrongs did not go to the Roigards’ house that day.
At 1.41 pm, Mr Roigard telephoned Ms Thoms on her cellphone. She answered and at Mr Roigard’s request handed the phone to Mr Bevans. Mr Roigard asked Mr Bevans if he had seen or heard from Aaron. Mr Bevans said that Mr Roigard sounded “quite upset”. On Mr Bevans’ evidence, the appellant told him there had been “a bit of confrontation” between Aaron and the people who, on the appellant’s account, had been selling the farm. Mr Roigard said he was not there but he had heard a “kerfuffle over at the house where they were supposed to be doing this paperwork”. He then told Mr Bevans that, “Basically Aaron stormed off out to the car, drove the car down to the end of the driveway and got picked up by a car”.
Mr Bevans also said that the appellant told him that the car in which Aaron drove away was a dark green or blue Holden Commodore. He said it had “skidded off quite fast” in the direction of the coast. Mr Bevans’ evidence was that when Mr Roigard told him what had happened, Mr Roigard was still upset and was close to tears. Mr Roigard provided further detail about the argument that had taken place. He said he heard Aaron say “You can shove your f…kin farm up your ass”. He said Aaron had said this to “a woman, to do with this farm”.
On Mr Bevans’ account, Mr Roigard then told him that he had run down the track in an unsuccessful attempt to catch Aaron. Aaron’s car[11] had been left at the end of the track. It was then that the appellant apparently called Ms Thoms’ phone and spoke to Mr Bevans. He told Mr Bevans he would bring the car back to Aaron’s place and then get a ride home. About 20 minutes later, Mr Roigard arrived in Aaron’s car which he parked. On Mr Bevans’ description the appellant was “shaking” and “almost crying” as he reiterated what he had already told Mr Bevans. Mr Bevans then drove the appellant home.
[11]We refer to this car as Aaron’s car although the evidence was that this was Ms Thoms’ car that Aaron often drove, and was driving on 2 June.
A search for Aaron began. At 4.45 pm that day, Mr Roigard sent a text message to Ms Thoms stating that the Armstrongs were “very f…king pissed” with Aaron for the way in which he had talked to them when they met. Ms Thoms saw that the farm to which they were meant to be moving was still occupied. She sent a text message to Mr Roigard indicating she wanted the truth and suggesting something was not right. There was no answer to that message and at 5 pm Ms Thoms reported Aaron missing. What began as a missing persons inquiry subsequently became a homicide inquiry. As has been noted, Aaron’s body has not been found.[12] Nor was the Holden Commodore, to which Mr Roigard had referred, sighted.
[12]The jury had evidence about the searches undertaken.
Mrs Armstrong’s evidence was that her last contact with Aaron would have been some years before the day he went missing. Mr Armstrong said he had seen him about four to six weeks before that day. Both Mr and Mrs Armstrong said they had not had any discussions with the appellant about investments or investing money, or with respect to helping with the purchase of a farm by Aaron. There was no challenge at trial to that evidence.
The Crown case against the appellant at trial
The Crown case was that Aaron had been murdered by the appellant “at some time after the 12.02 pm phone call” and before a text message saying “They on there way” which “the Crown claimed the appellant had sent to Ms Thoms from Aaron’s phone at 12.12 [pm]”.[13] Mr Roigard had disposed of Aaron’s body “at a pre-arranged location” after that.[14] As the Court of Appeal also noted, the case was put on the basis that the appellant “had planned to carry out the killing, motivated by his desire to cover up his fraud on Aaron”.[15]
[13]CA judgment, above n 3, at [21].
[14]At [21]. The Crown in closing suggested Mr Roigard had about an hour and a half in which to do so, fixing that time by reference to the call Mr Roigard made to Ms Thoms at 1.41 pm.
[15]At [21].
As this excerpt from the Court of Appeal judgment indicates, the evidence of the appellant’s fraud in relation to the investment account was advanced by the Crown as motive and context for Aaron’s murder. The evidence in the Crown case had a number of other strands which can be summarised briefly.[16]
[16]The relevant evidence is summarised by the Court of Appeal at [25]–[31]. This summary also draws on the written submissions for the respondent.
First, there was evidence to suggest that Mr Roigard had Aaron’s cellphone after the time when he told police that Aaron had disappeared. The cellphone has not been recovered and was not able to be activated since just after 1.02 pm. A number of matters were relied on in this respect. These matters included evidence suggesting text messages sent from the cellphone between 12.12 pm and 12.55 pm had been authored by Mr Roigard. There was also evidence Mr Roigard told Ms Thoms that Aaron had received an unanswered call from a David Wright at 1.02 pm, which he could only have been aware of if he had access to Aaron’s cellphone at that time.[17]
[17]There was also evidence from a man who had met Mr Roigard some years earlier and who had rung Mr Roigard on reading about Aaron’s disappearance. He said Mr Roigard told him that Aaron had left his cellphone and his EFTPOS card behind when he disappeared.
The next strand of the Crown case comprised the evidence from witnesses from the Institute of Environmental Science and Research Ltd (ESR) of the forensic inquiries undertaken at the Roigard farm. We come back to the detail of some of this evidence but it is sufficient to note here that there was evidence about what was identified as blood in an area of just over a metre by a metre between two woodsheds on the property.[18] On the appellant’s account to police, he and Aaron were in this area on 2 June and there was evidence to suggest that the two men had been working with firewood on Sunday 1 June, the day before Aaron’s disappearance.
[18]On 12 June 2014 ESR had undertaken tests with luminol, a chemical which reacts with blood.
Further, the ESR found six stains of blood on the door of one of the woodsheds. Samples from these stains were consistent with them being Mr Roigard’s blood.
A wood splitter, described as very rusty and blunt, located in one of the woodsheds was found to have two tiny stains of blood on the handle.[19] A partial DNA profile was obtained from one of the stains which was consistent with it being Aaron’s blood. The ESR witness, Glenys Knight, thought it “highly unlikely” that this wood splitter was the murder weapon and the Crown did not suggest it was. Rather, the Crown contended the murder weapon was another wood splitter which had been buried along with Aaron’s body.
[19]The appellant’s neighbour explained that a wood splitting axe was “a lot heavier [than an axe and has] a very heavy type head on it but still has the blade”. The neighbour also explained he used a hydraulic splitter attached to his small tractor to split wood.
The next strand in the Crown case comprised evidence which suggested the appellant was away from his property at a time consistent with the disposal of Aaron’s body. This evidence was inconsistent with what the appellant had told the police. This evidence included polling data about a call made by Mr Roigard to Aaron’s voicemail at 1.14 pm which suggested it was not made from the property and a message sent to Ms Thoms from Aaron’s cellphone at 12.47 pm which on the Crown case was sent by the appellant.
The Crown also relied on evidence from a friend of the appellant, Phillip Hopkinson, about contact with Mr Roigard on 9 May 2014. Mr Hopkinson said a distraught Mr Roigard explained that Aaron had been missing for two days, having left without his car or money. Aaron was not in fact missing at this time. Then, on the evening of 2 June after Aaron had disappeared, Mr Hopkinson said the appellant was crying and told him Aaron was “still missing”. The Crown attached some significance to the timing of the May “disappearance”. That was because of the coincidence in timing as against Aaron’s potential farm purchase.
There was also evidence about the undulating topography of the area which the Crown relied on to suggest a body could have been buried or hidden from sight.
Finally, in terms of the circumstantial aspects, the Crown pointed to various inconsistencies in the appellant’s numerous accounts of the circumstances of Aaron’s disappearance. The jury had evidence of oral and written statements made by Mr Roigard to police, as well as a videotaped interview between Mr Roigard and police and a videotaped reconstruction of events as Mr Roigard said they unfolded. The defence acknowledged Mr Roigard was an inveterate liar.
As noted, the Crown also called evidence from Mr F and Mr W, both of whom described conversations they said they had with Mr Roigard when they met him in prison.
The inmate evidence
Mr F explained in his evidence that he had met Mr Roigard when they were in the dayroom together whilst Mr Roigard was on remand in Kaitoke Prison in November 2014. He described a number of conversations with Mr Roigard which took place on 1, 2, and 9 November and then later, in a conversation which he said took place around the end of the year, or possibly early the following year.
Mr F said that Mr Roigard initially told him that the police case against him was weak and that they were looking on his computer and checking the polling on the cellphone. Mr F said they had a discussion about the money (about $68,000) that Aaron had paid through to the account. He said that Mr Roigard mentioned the word “Sovereign” and that the money had led to a rift between the appellant and Aaron.
Mr F said that Mr Roigard told him that the police had overlooked some things, particularly, a splitter which had only been taken from the property at a later stage. Mr F thought the splitter was an axe for cutting wood.
In the second of their conversations, Mr F said that Mr Roigard told him that police had found his blood somewhere on the farm. He said that Mr Roigard told him he had “picked up the splitter and lashed out with it and hit Aaron” three times over the head while Aaron was walking away. Mr F’s evidence was that Mr Roigard said that this had occurred at the same place where Mr Roigard’s blood had been found. Mr F said that Mr Roigard told him that he had cleaned up the site with “a scoop or something of that nature” and somehow moved the body. In a later conversation, Mr F said he asked Mr Roigard why he had not told the police that this was a “spur of the moment” thing. His evidence was that Mr Roigard told him he did not think they would believe him because of the circumstances and the money.
Mr F also said they had a discussion which Mr F thought was about where the body might be placed, the comment being “A long way up Eltham Road”. In another discussion Mr Roigard said that it was deep, and it was on the top of Hastings Road. (A search of this area was undertaken by police.)
Mr F was cross‑examined about his criminal history. He accepted that since 1981 he had amassed 154 convictions of which 138 were dishonesty offences. He was asked in some detail about his most recent fraud offending (following a Serious Fraud Office prosecution) and about his escape from custody which culminated in him restraining and threatening his wife with a weapon.[20] It was put to him that what he had said in his evidence at trial about Mr Roigard cleaning things up with a scoop was inconsistent with his earlier statement to police in which he had referred to the scoop being on a tractor. He was also questioned about variations in the type of vehicle he said that Mr Roigard told him he had used to transport the body. Finally, Mr F was cross-examined about the credit by way of a discount in sentence, in respect of his conviction for escaping custody and kidnapping, which he had received for assistance to police. As the Court of Appeal noted, he also “reluctantly conceded” that his assistance might be something relied on at an appeal which was to take place the following week “but professed disappointment at the way he had been treated by the police” and referred to his concerns for the security of himself and his family.[21]
[20]It was after this incident that he was apprehended and put in the At Risk Unit at Kaitoke Prison.
[21]CA judgment, above n 3, at [53]. There was no adjustment to sentence on appeal: [F] v The Serious Fraud Office [2016] NZHC 271.
Mr W’s evidence was not as extensive. He also said he had met Mr Roigard while they were both remanded in custody.[22] Mr W told the jury that Mr Roigard had said they would never find the body, that “they’d never find his son” and “that he got what he deserved”. Mr W also explained that Mr Roigard told him about the cellphones and “just bits and pieces”. He said that Mr Roigard had talked about an axe with a speck of Aaron’s blood on it.
[22]Mr W was remanded in Kaitoke Prison in early August 2015. He was transferred to the same wing as Mr Roigard in September 2015. It was in this wing that Mr W said he came into contact with Mr Roigard.
Mr W went on to say that Mr Roigard told him that he did not know where Aaron was and that Mr Roigard stated that he would look for Aaron if he got out of prison. Mr W also told the jury that Mr Roigard said that he and his son had had an argument and that they then went in his son’s car. Mr Roigard then came back and left the car at the end of the driveway. His evidence was that he learnt from Mr Roigard that Aaron’s partner had found out that the “farm thing” was not true. Mr W said that he thought that Mr Roigard was supposed to be saving Aaron’s money but that it was not there. His evidence also included the observation that Mr Roigard told him that Mr F was testifying to get time off his sentence and that Mr F was lying.
Mr W was cross‑examined about the offences for which he was in prison and about his other convictions as well as a discount on sentence he received. He accepted that, between 1992 and October 2015, he had 112 convictions of which 64 involved dishonesty. He also accepted he had contacted police saying he had some information about Mr Roigard in late September 2015.
After the two inmates gave their evidence, the Judge directed the jury about the benefits that can be received for assistance of this sort. A similar direction was included in the Judge’s summing up which also explained that both witnesses had a motive to give false evidence because of their expectation of a benefit.
The defence case at trial
The defence case had three main limbs. The first of these was that the defence said there was an insufficient basis on which the jury could be sure the Crown had proved that Aaron was dead. Second, if the jury was sure he was dead, the defence advanced suicide as a possible explanation for Aaron’s disappearance. Third, if the jury believed that Mr Roigard had killed Aaron, the defence contended that, on that assumption, the evidence did not support the Crown case of his planned and premeditated murder by Mr Roigard. The more likely proposition was, on the defence theory, that if the jury found Mr Roigard had killed his son then this was manslaughter. The defence relied in this respect on the evidence of Mr F.
The defence challenged the credibility of both Mr F and Mr W. It was accepted that conversations with the two men took place and that those conversations canvassed the case against Mr Roigard. However, it was not accepted that Mr Roigard admitted to killing Aaron or to hiding his body.
The Court of Appeal decision
Four grounds of appeal were raised in the conviction appeal in the Court of Appeal. The first of these was a challenge to the admissibility of the evidence of Mr F and Mr W. The Court of Appeal considered that the appellant’s argument that this evidence was inadmissible was based on the “generic characteristics of inmate confession evidence”.[23] The Court said that this argument could not succeed because it was contrary to this Court’s decision in Hudson v R that there was no presumption that evidence of prison informants like that of Mr F and Mr W was inadmissible.[24]
[23]CA judgment, above n 3, at [41].
[24]Hudson v R [2011] NZSC 51, [2011] 3 NZLR 289.
The second ground of appeal related to the Judge’s directions about the inmate evidence. The Court considered that, when looked at in context, the directions fulfilled the requirements. Next the Court dealt with the third ground of appeal: whether trial counsel, Mr Keegan, had acted in accordance with instructions when running the manslaughter defence. The Court concluded that he had. Finally, the Court addressed the fourth ground of appeal, which was a challenge to the admissibility of other evidence. This included evidence that Mr Roigard accessed a Yahoo site on 10 May 2014, the day after he told Mr Hopkinson that Aaron was missing, addressing the question “Why can one blow to the head kill you instantly?” The Court considered this evidence was properly admissible.[25]
Exclusion under s 8 of the Evidence Act
Context
[25]Leave to appeal against the Court of Appeal’s decisions on the second, third and fourth grounds was declined: Leave judgment, above n 4, at [4]–[7].
Before turning to the submissions made about s 8, it is helpful to put the discussion in context by providing a brief description of the relevant sections of the Evidence Act and of the approach taken to the evidence of prison informants by this Court in Hudson.
We begin with s 6(b) which states that the purpose of the Act “is to help secure the just determination of proceedings by … providing rules of evidence that recognise the importance of the rights affirmed by the New Zealand Bill of Rights Act 1990”. Under s 10(1)(a), the Act is to be interpreted in a way that promotes that purpose.
Section 7(1) sets out a “fundamental principle” that all relevant evidence is admissible unless inadmissible or excluded under the Evidence Act or any other Act. Under s 7(2), irrelevant evidence is not admissible. Evidence is relevant “if it has a tendency to prove or disprove anything that is of consequence to the determination of the proceeding”.[26]
[26]Evidence Act 2006, s 7(3).
Section 8 describes the principles of “general exclusion”. Section 8(1)(a) provides for the exclusion of evidence where its probative value is outweighed by the risk that the evidence will have an unfairly prejudicial effect on the proceeding. In deciding whether the probative value is outweighed in this way, “the Judge must take into account the right of the defendant to offer an effective defence”.[27]
[27]Section 8(2).
The applicability of ss 7 and 8 to evidence from prison informants was considered in Hudson. In Hudson it was accepted that “the evidence of admissions allegedly made by the [appellant] while in prison to other prison inmates requires careful scrutiny”.[28] The Court said that this type of evidence was not presumptively inadmissible, but the Court also recognised:[29]
… that there may be scope for excluding prison admission evidence under ss 7 and 8 of the Act, but, that said, the legislative scheme as a whole is indicative of a legislative intention that reliability decisions ought to be made by a properly cautioned jury.
[28]Hudson, above n 24, at [33].
[29]At [36].
We declined to revisit Hudson on this appeal.[30]
The submissions
[30]Leave judgment, above n 4, at [3].
The appellant says that the evidence of Mr F and Mr W should have been excluded under s 8(1)(a) on the basis that any probative value was outweighed by the risk of illegitimate prejudice.
In developing the submissions on this point, the appellant first addresses the implications of Hudson. Three main arguments are made. The first of these is that Hudson has been misinterpreted because of the emphasis incorrectly placed on the determination that there is no presumption of inadmissibility. The appellant emphasises that does not equate to a presumption of admissibility. Second, it is contended that the conclusion that the admission of the informant evidence in Hudson did not give rise to a miscarriage of justice has to be considered in light of the fact that the jury directions about the informants’ evidence in Hudson were much stronger than those in the present case. That meant the directions were more responsive to the prejudicial impact of this evidence. Finally, the appellant submits that the evidence of Mr F and Mr W requires the “careful scrutiny” envisaged by Hudson given the indicia of a lack of probative value.
In that context, the appellant submits that reliability considerations form part of the assessment of probative value under s 8. Here, there are “serious and obvious concerns about [the] reliability or the credibility of the witnesses” who are “classic” prison informants. In this respect, it is noted that Mr F and Mr W both knew about the system of incentives available for assistance to police and both received them. The appellant also advances the submission, expanded upon in the context of s 30, that it is not possible to buy a witness and that the provision of incentives operates to have the effect of doing so. It is further argued that the ability of the Crown to “reward” witnesses in this way operates unequally against defendants who cannot do the same and is inconsistent with rights in the Bill of Rights Act including the right to a fair trial and to equality of arms in relation to the examination of witnesses.[31]
[31]New Zealand Bill of Rights Act 1990, s 25(a) and (f).
In support of the arguments about the need to address reliability in assessing probative value, the appellant relies on the social science literature about miscarriages of justice which have resulted from the use of evidence like that in issue in this case. Counsel emphasises the studies which identify difficulties juries have in assessing the impact of incentives on the evidence of prison informants and the tendency of juries to overvalue this evidence.[32]
[32]On these and other matters, counsel adopts the submissions for Mr W in W (SC 38/2019). The relevant social science literature is discussed in more detail in W (SC 38/2019), above n 6,In addition, it is submitted that admission of the evidence risks illegitimate prejudice because of the risk its presence creates of the jury reasoning in what Mr Lithgow QC describes as a “superficial or emotional way”. That submission is linked to the argument that the evidence from Mr F and Mr W would have served to confirm the jury’s suspicions that Aaron’s disappearance resulted from murderous actions by Mr Roigard.
In terms of Mr F, Mr Lithgow says that his ability to be a reliable witness is severely impeached. He refers to Mr F’s history of dishonesty offending and to statements from a number of judges in other contexts about Mr F’s long-term dishonesty. It is also submitted that Mr F offered to give information about Mr Roigard on the basis that he would obtain a benefit. The evidence of Mr W is seen as raising similar concerns.
For the respondent the submission is that the evidence was properly admitted under s 8. It is accepted that in an extreme case where the evidence is incapable of belief or where the evidence defies belief, the judge may exclude it. Apart from that, the submission is that questions about the reliability or credibility of the evidence of the kind in issue in this case were for the jury. In addition, Ms Markham makes the point that the appellant’s criticisms ignore the evidence at trial. The submissions highlight consistencies between the other evidence at trial and that of Mr F and Mr W. It is also argued that the evidence of Mr F was relied on by the appellant to support the manslaughter defence run at trial.
Principles
The relevant principles about reliability considerations in the s 8 assessment of the evidence of prison informants are set out in the other judgment being delivered today, W (SC 38/2019).[33] Some adjustment in the consideration of those principles is necessary to reflect the fact that W (SC 38/2019) is a pre-trial decision whereas here we have the benefit of the evidence at trial. With that qualification, it is helpful to re‑state the approach taken in W (SC 38/2019) in which the Court has noted that reliability considerations are relevant in evaluating the admissibility of the evidence of prison informants under s 8 (and under s 7).[34] The evaluation to be made under s 8 in W (SC 38/2019) in relation to the admission of the evidence of prison informants in summary is as follows:[35]
(a)The concern in undertaking this evaluation is to determine “whether the connection between the evidence and proof is ‘worth the price to be paid by admitting it in evidence’”.
(b)In undertaking the gatekeeping role, reliability may be considered by the judge in balancing, in the usual way, the probative value of the proposed evidence against the risk of illegitimate prejudice. That reliability assessment should be made without applying any artificial limits or presumptions such as taking the evidence at its highest.
(c)The relevant factors will include consideration of the sorts of concerns about this evidence as have been discussed [with reference to the social science research material] and which might include, for example, that the credibility of the witness in an informant context has previously been doubted, any incentives or expectations of preference at play (including the inability of the prosecution to confirm whether incentives have been offered or given), and the likely weight to be attached to this evidence.[[36]]
(d)On the other hand the exercise is not that of a mini trial. The judge will be making his or her assessment in the absence of the full picture of the evidence as it may emerge at trial.
(e)Finally, the constitutional role of the jury as fact‑finder needs to be respected. … [T]he statutory scheme and the authorities, particularly Hudson, which we are not overruling, envisage that the court will utilise other mechanisms such as clear judicial directions to the jury to address the generic risk of unfair prejudice.
[33]W (SC 38/2019), above n 6, at [87]–[89].
[34]Mr Lithgow made some references to s 7 but this was not the focus of the case so we say nothing further about relevance.
[35]At [88] (footnotes omitted).
[36]As noted in W (SC 38/2019), above n 6, at [88(c)], n 141, the usual, more generic factors, such as a history of dishonesty offending and any animus towards the defendant, will also be part of the equation.
We add, as we did in W (SC 38/2019),[37] that we do not consider the further formulation of the framework set out in the reasons of Winkelmann CJ should be used.[38] The application of that framework in this case has resulted in an approach which requires independent corroboration of the evidence in issue and which places emphasis on the need for the court in a case such as this one to ask whether the evidence of a confession has been constructed by the witness to cohere with facts they have gained from other sources.[39] We see these aspects as matters for trial and cross‑examination. In addition, while the presence of independent corroboration is a relevant factor, it should not be elevated to a requirement for admission. Finally, as we said in W (SC 38/2019),[40] we make it clear that we leave the approach to the evidence of other witnesses who are not prison informants, but who may be incentivised in some way in the context of the criminal justice system, to be determined as cases involving that evidence arise.
[37]At [87].
[38]See the reasons of Winkelmann CJ in W (SC 38/2019), above n 6, at [253]–[270] and below at [108]–[114].
[39]As we discuss, we also consider that in this case whether there was a plausible narrative that Mr F constructed his evidence from details Mr Roigard gave him of the police case was a matter for the jury.
[40]At [87].
Counsel in this appeal and in W (SC 38/2019) were asked to provide further submissions on the current safeguards that apply to the admission of the evidence of prison informants in addition to those provided by the judge in exercising the gatekeeping role under ss 7 and 8. As we noted in W (SC 38/2019), our review of the material from the parties in both cases, together with the social science material relied on in both cases, suggests there is a need for additional safeguards.[41] In W (SC 38/2019) we discussed the need for further guidance for prosecutors and for the maintenance of a central register of those who have given evidence as a prison informant and of how that evidence was treated.[42] This will include a record of the criminal history of those who give evidence in this way and capture any reductions in charge or in sentence or any other preference or benefit gained for any assistance provided.
[41]At [90]–[96].
[42]At [93].
We turn now to apply these principles to the evidence of Mr F and Mr W.
Assessment
Both Mr F and Mr W are classic prison informants. Their evidence is solely derived from their knowledge of, and conversations with, Mr Roigard in prison. Both were aware of the potential benefit they might obtain from giving their evidence and both received a sentencing discount for assistance provided. And, as has also been noted above, both have a history of dishonesty.
In terms of the incentives received, Mr F had approached police in early November 2014 and we were told he had provided two written statements.[43] His approach came not long after he was sentenced to a term of imprisonment of three years and one month in respect of the offending prosecuted by the Serious Fraud Office. After a sentencing indication, he ultimately received a sentence of two years’ imprisonment for the escape from custody and kidnapping which was cumulative on the sentence for the fraud offending.[44] That sentence reflected a discount for assistance in this case.[45]
[43]One in November 2014 and the other in March 2015.
[44]R v [F] [2015] NZDC 11167.
[45]The total discount given for mitigating factors including guilty pleas and assistance was 50 per cent from a starting point of four years.
When Mr W was talking with the appellant, he was on remand on charges of dealing methamphetamine, receiving, and unlawful possession of firearms. He received a discount at sentencing on 24 November 2015 for his assistance and for assistance in an unrelated matter which resulted in police recovering firearms.[46]
[46]The discount equated to just over 52 per cent and also reflected guilty pleas. It appears the discount for the prospective testimony in the present case was greater than that given for the assistance leading to the recovery of firearms: R v Roigard [2015] NZHC 3163 at [9].
Accordingly, the evidence of Mr F and Mr W directly raises the concerns about incentivised prison informant evidence discussed in the social science literature and, in particular, the risks that this evidence is overvalued despite the motivation to present false evidence.[47]
[47]See W (SC 38/2019), above n 6, at [76]–[85]. The arguments for the respondent, repeated in this case, about the limitations of the social science literature are addressed in that discussion.
On the other side of the equation, the following points can be made.
First, it is not disputed that these conversations took place with both witnesses and that the case against Mr Roigard was discussed. Significant portions of the evidence are not challenged.[48] With Mr F the only real issue, apart from some questions about peripheral matters, is as to the admission that Mr Roigard hit Aaron three times and hid the body. Obviously that is the critical evidence but it would be artificial here to look at that part of the conversation in isolation. The same point applies to the evidence of Mr W.
[48]We see no reason to discount this indication of reliability.
Further, large parts of Mr F’s evidence, as Ms Markham submitted, conform with the evidence given at trial by other witnesses. The forensic evidence, which indicated the presence of blood around the area where Mr F said Mr Roigard told him that he had hit Aaron, provides an illustration.
The ESR evidence was that four of the stains of blood found on the door of one of the woodsheds were contact or transfer stains and two were landed stains.[49] An ESR witness explained the difference between a “transfer” stain and one which has “landed” in this way:
… contact or transfer blood stains are stains that have occurred via contact with a source of blood. So if a blood stained surface contacts another surface some of that blood will transfer if it’s wet and that’s what we refer to as a transfer blood stain. The other [type] of blood stain is blood that has travelled through the air by some mechanism and landed on a surface, so whether it’s been flicked or coughed or spattered by some other mechanism but it’s something that’s actually travelled through the air and landed on a surface.
[49]See above at [19].
The tiny stains of blood identified on the handle of the wood splitter as consistent with Aaron’s blood[50] were “spatter” stains. The jury were told that “spatter” stains described something where the blood “has come through the air and landed on there, it’s not been transferred by another blood stained item”. The appearance of the stains meant a source of wet blood had been subject to a “fairly forceful action” of some type. In re-examination, the ESR witness suggested the most likely cause for spatter of the size in question was “a quite strong force applied to a stationary pool of blood or area of blood”.[51] This evidence should not be over-emphasised because it does not exclude other possible explanations for the presence of blood on the handle. But that does not detract from the fact that the police only found the wood splitter through Mr F and the jury could have viewed it as generally consistent with Mr F’s narrative of events.
[50]See above at [20].
[51]One of the ESR witnesses who undertook the investigation on the property on 12 June 2014 indicated that the effect of the rain in the area in the period after Aaron’s disappearance would have diluted blood in the area and washed it down into the soil so it could no longer be seen.
Mr Lithgow suggests that Mr F could have made some assumptions about the nature of the death from what Mr Roigard did tell him but that possibility does not account for all of the detail he provides.
It is the case that what Mr F said about when police located the wood splitter cannot be right.[52] That is because the evidence is that the police did not pick up the splitter until well after the conversations between Mr F and Mr Roigard took place. But given the absence of evidence as to any source of this material, other than Mr Roigard, the jury could have inferred that this is what Mr Roigard told Mr F. It was not suggested to Mr F that police provided Mr F with any information by which he might have “flowered up” or embellished his account of these matters. And Mr Roigard’s narrative could be treated as corroboration. Nor was there evidence that he obtained information about the case from any source other than Mr Roigard, although the jury was directed to consider that possibility and whether the evidence of Mr F (and Mr W) could be “independently confirmed” by other evidence it accepted.
[52]See above at [28].
Mr Lithgow also submits there is reason to be sceptical about the Crown case that another wood splitter may have been the murder weapon and had been buried with Aaron. Obviously, the jury were not able to speculate about that but on Mr Roigard’s account, he and Aaron were in the area of the woodsheds on 2 June.[53]
[53]There was evidence police had not found any other axe-like tools or wood splitters on the Roigard farm. The appellant’s neighbour explained he was originally going to help the appellant move logs in the paddock outside the appellant’s woodshed on the Monday that Aaron disappeared but his chainsaw was broken and could not be fixed in time. He arranged to pick the logs up the following Saturday.
Against this background, ultimately, in our view whether the disputed part of Mr F’s account is credible and reliable were matters for the jury who did have the full picture about Mr F’s incentives and history of dishonesty. In these circumstances, there was no illegitimate prejudice from the admission of this evidence. The bulk of the evidence was not challenged and the evidence fitted with other evidence at trial. These are the key features in our assessment of where the balance lies under s 8.
Mr W’s evidence does not add much as it transpired. It appears that Mr W in his evidence did not come fully up to brief but was rather, as Ms Markham put it, “a somewhat unimpressive witness”. He mumbled at times and, as he said, his memory was “not the flashest”. And he was reluctant to accept responsibility for some of his earlier dishonesty offending. Finally, at one point, he said he did not “really like believe half of what [Mr Roigard] said anyway” and went on to say that “you could never tell whether he was just fantasising or telling the truth, … just dunno, bizarre stories, yeah, like…”. That said, his evidence does have some probative value because of the reference to the fact Aaron would not be found and “that he got what he deserved”.
In terms of assessing reliability, his evidence largely conforms with other evidence led at trial. An illustration of this is his evidence that the appellant said that he and Aaron went for a drive. Mr W was not cross-examined about this and it gels with the evidence of Jack Scott who said he saw Aaron in his car with another person, who he thought was a male, late in the morning of Monday 2 June, the day Aaron disappeared. This was not consistent with the appellant’s account to police in which he said they stayed on the farm until Aaron left and that Aaron parked his car at the Roigard farm. The questions about the reliability and credibility of this evidence were for the jury and the probative value of the evidence favoured its admission in terms of the s 8 analysis.
The application of s 30 of the Evidence Act
Section 30 sets out the basis on which “improperly obtained” evidence is excluded. In this case, the appellant relies on s 30(5)(c) which provides that evidence which is obtained unfairly is improperly obtained.
As we have noted, Mr Lithgow’s proposition is that a witness cannot be bought. He says that if evidence is bought from someone like Mr F and Mr W, via the provision of a sentence discount, that evidence is obtained improperly in terms of s 30. He emphasises that time, in the form of a sentence discount, is an important commodity in this respect. In developing this submission Mr Lithgow says that obtaining “unfairly” under s 30(5)(c) includes the way in which the system as a whole operates.
Reprising submissions made in the Court of Appeal, Mr Lithgow challenges the recognition courts provide for assistance, such as that provided in this case, by giving sentencing discounts. He says in this way the courts have approved a system, the public good of which is untested. It is also submitted that the courts have an inherent or implied power to reject evidence obtained in this way.[54]
[54]Reference is made to s 11 of the Evidence Act which states that “[t]he inherent and implied powers of a court” are not affected by the Act, “except to the extent that [the] Act provides otherwise”.
The respondent reiterates the submission made on this aspect in W (SC 38/2019), namely, that s 30 is directed to the improper or unfair conduct of officials in the obtaining of evidence and, as well, to circumstances involving search, interrogation, entrapment and undercover exercises. Ms Markham also refers to the Practice Note – Sentencing 2003 which sets out the obligations of Crown and defence counsel applicable to those sentencing decisions in which credit may be given for assistance to police,[55] and to the case law recognising that practice. Against that background, she questions the basis on which that process can be described as unfair. Finally, the submission for the respondent is that it would be odd to exclude evidence under s 30 when it formed a key part of the defence strategy at trial, that is, the manslaughter defence.
[55]Practice Note – Sentencing 2003 [2003] 2 NZLR 575 at [4.1]–[4.2]. See also R v Hadfield CA337/06, 14 December 2006 at [15]; and Hessell v R [2009] NZCA 450, [2010] 2 NZLR 298 at [23].
Taking the last point first, if the evidence in issue was improperly obtained, the fact the defence utilised the evidence in support of the manslaughter defence does not provide an answer to concerns as to the admissibility of the evidence. Such an approach would not be consistent with the Evidence Act nor with the proper administration of justice.[56] Once the evidence was admitted, the defence was entitled to make the best of that evidence. We add there was no evidence to suggest that the failure to challenge the admissibility of the evidence in the first place was a tactical decision by the defence.[57]
[56]For the same reason, to the extent that the defence use of the evidence was relied on by the respondent in support of the submission that the evidence was reliable under the s 8 analysis, it is rejected.
[57]Trial counsel, Mr Keegan, in his affidavit filed in the Court of Appeal said he was troubled by this evidence but his instructions were that the conversations took place except that Mr Roigard said he did not confess to assaulting or injuring Aaron or hiding the body. Mr Keegan said he did not consider the Court would edit the statements to reflect those instructions.
Addressing then the matters raised by the appellant, there is authority which challenges the proposition that witnesses cannot receive payment for giving their evidence.[58] In addition, the practice of giving credit for assistance in the form of evidence for the Crown or for the provision of other assistance to the Crown dates back in the United Kingdom to at least 1913.[59] It is a recognised practice in other comparable jurisdictions[60] and in some jurisdictions there is statutory recognition of the approach.[61] However, we need not reach any concluded view on either of Mr Lithgow’s propositions and can leave the question of the appropriateness of providing credit by means of sentence discounts for assistance provided to another day. Rather, this part of the case can be decided on the same basis as was applied in W (SC 38/2019).
[58]For example, R v Kennedy [2004] 3 NZLR 189 (CA) where a witness in a Serious Fraud Office investigation was paid for providing assistance. In Hudson there was a reward for the provision of information about the murder: Hudson, above n 24, at [12]. See also R v Chignell [1991] 2 NZLR 257 (CA); Mark Lucraft (ed) Archbold: Criminal Pleading, Evidence and Practice (2020 ed, Sweet & Maxwell, London, 2020) at [5A-91]; and Sidney N Lederman, Alan W Bryant and Michelle K Fuerst Sopinka, Lederman & Bryant: The Law of Evidence in Canada (5th ed, LexisNexis, Markham (Ont), 2018) at §17.12, citing R v Dikah (1994) 18 OR (3d) 302 (CA) (affirmed in R v Dikah [1994] 3 SCR 1020).
[59]The respondent refers to R v James (1914) 9 Cr App R 142 (Crim App) at 144 as an example of the history of this practice. See the discussion of James and other authorities in Geoff Hall “Sentencing (II): Matters of aggravation and mitigation” [1985] NZLJ 184 at 189; Geoff Hall Hall’s Sentencing (online ed, LexisNexis) at [I.7.2]; and R Paul Davis “Sentencing the Informer” (1980) 144 JPN 249 at 249–250. See also the discussion of the common law practice in Lucraft, above n 58, at [5A-104]–[5A-110].
[60]For the position in Canada, see generally Clayton C Ruby and others Sentencing (9th ed, LexisNexis, Toronto, 2017) at §§5.128–5.140.
[61]In England, Wales and Northern Ireland see Serious Organised Crime and Police Act 2005 (UK), s 73. See also the discussion of s 73 in Lucraft, above n 58, at [5A-92]–[5A-103]. In Australia, there is direct statutory recognition of this practice in pieces of federal and state legislation: see for example Crimes Act 1914 (Cth), s 16A(2)(h); Crimes (Sentencing Procedure) Act 1999 (NSW), s 23; Sentencing Act 2017 (SA), s 37; Crimes (Sentencing) Act 2005 (ACT), ss 33(1)(l) and 36; Sentencing Act 1995 (NT), s 5(2)(h); and Penalties and Sentences Act 1992 (Qld), s 9(2)(i) and ss 13A–13B. The practice is also recognised, albeit less directly, in the Sentencing Act 1991 (Vic), s 5(2AB) and in the Sentencing Act 1995 (WA), s 8(5). See also Halsbury’s Laws of Australia (1995, online ed) vol 9 Criminal Law at [130-17135].
As in W (SC 38/2019), we do not need to resolve in this case the question of the exact scope of s 30 and, in particular, whether it is directed only to matters such as the fruits of search and other interrogation techniques as the respondent argues. Nor do we need address the argument based on the Court’s inherent and implied powers which relies on s 11 of the Evidence Act. We say that because, as in W (SC 38/2019), on analysis, the appellant’s arguments are premised on a presumption this type of evidence is inadmissible. The argument is ultimately dependent on the concept of a system which it is said inevitably operates unfairly so as to render this evidence improperly obtained. On Mr Lithgow’s approach it is difficult to see how any distinction could ever be made between admissions from prison informants that are properly obtained and those which are not. In other words, this is an argument for a presumption of inadmissibility. That presumption was rejected in Hudson and we are not revisiting that decision. Section 30 does not add anything in this case.
Result
For these reasons, in accordance with the view of the majority, the appeal is dismissed.
WINKELMANN CJ AND WILLIAMS J
(Given by Winkelmann CJ)
Table of Contents
| Para No. | |
| Introduction | [80] |
| Background | [83] |
| The Crown case | [83] |
| Treatment of evidence of prison informants at trial | [89] |
| The trial Judge’s direction | [97] |
| Court of Appeal | [98] |
| Argument on appeal | [104] |
| Framework for analysis | [108] |
| Admissibility of evidence of Mr F | [115] |
| Relevant background | [115] |
| Mr F’s evidence at trial | [115] |
| Mr F’s criminal history | [122] |
| Benefits received by Mr F | [125] |
| Application of framework | [127] |
| Probative value | [127] |
| (a) Significance of evidence to matter at issue | [128] |
| (b) Does Mr F have a record of lying? | [129] |
| (c) Was the evidence incentivised? | [135] |
| (d) Indications the evidence is unreliable or untrue | [138] |
| Unfair prejudice | [155] |
| Assessment | [159] |
| Admissibility of evidence of Mr W | [162] |
| Relevant background | [162] |
| Mr W’s evidence at trial | [162] |
| Mr W’s criminal history | [168] |
| Benefits received by Mr W | [169] |
| Crown reliance on Mr W’s evidence | [170] |
| Application of framework | [171] |
| Probative value | [171] |
| (a) Significance of evidence to matter at issue | [171] |
| (b) Does Mr W have a record of lying? | [172] |
| (c) Was the evidence incentivised? | [173] |
| (d) Indications the evidence is unreliable or untrue | [174] |
| Unfair prejudice | [175] |
| Assessment | [176] |
| Section 30 of the Evidence Act | [177] |
| Conclusion | [178] |
| Outcome on appeal | [179] |
Introduction
Mr Roigard was convicted of the murder of his son, Aaron. On this appeal, he argues that a miscarriage of justice has occurred because at his trial the jury heard evidence from two witnesses of conversations they claim to have had with Mr Roigard in prison, evidence which he says should have been excluded.
Mr Lithgow QC for Mr Roigard argues that for personal gain – the reduction of jail time – two witnesses, Mr F and Mr W, each lied when they claimed that, in conversation with them, Mr Roigard made admissions suggesting he had killed his son. These were, Mr Lithgow says, classic prison informant witnesses giving evidence that is easy to manufacture and impossible to refute.[62] The witnesses were incentivised to give evidence against Mr Roigard. They were in substance “paid” in the sense that, in prison, time is the most valuable currency there is.
[62]Impossible to refute unless the defendant is able to show that there was no opportunity for the conversation in question. See the discussion in Hudsonv R [2011] NZSC 51, [2011] 3 NZLR 289 at [18].
The approved question on appeal is whether the Court of Appeal erred in upholding the admissibility of the evidence of these witnesses. As is the case in W (SC 38/2019) v R (W v R),[63] judgment in which is being delivered at the same time, leave was granted on the basis that the Court will not revisit its decision in Hudson v R.[64] In Hudson, the Court held that there is no presumption of inadmissibility of prison informant evidence.[65]
Background
The Crown case
[63]W (SC 38/2019) v R [2020] NZSC 93 [W v R] (details of which are suppressed until final disposition of trial).
[64]Roigard v R [2019] NZSC 63 [Leave judgment] at [3]; and Hudson, above n 62.
[65]At [36].
The Crown case at trial is set out fully in the reasons of the majority and we do not repeat it here; rather, we highlight aspects of the case we consider relevant to the issues on this appeal.
Without the evidence of Mr F and Mr W, the Crown case against Mr Roigard was entirely circumstantial. This is not to say it was a weak case. The evidence strongly suggested that Aaron was dead and that Mr Roigard was the last person to see him alive. There was also compelling evidence that Aaron’s disappearance occurred against a background of conflict between father and son. Mr Roigard had been managing a savings scheme for Aaron for several years so that Aaron could buy a farm. In reality, Mr Roigard had been using the money for his own purposes so that there were no savings. Aaron’s resolve to go ahead and buy his own farm was set to expose Mr Roigard’s theft on the very day of Aaron’s disappearance.
On the morning in question, Aaron left his home and drove to his parents’ farm. He had arranged with his father to meet there to sign papers to complete a farm purchase he believed his father had organised for him, utilising the savings he believed his father held for him. In reality, no purchase had been arranged. The Crown had evidence that tended to prove Mr Roigard lied about what occurred on that morning and in particular about his and Aaron’s movements around the time the Crown alleges Aaron was killed.
Mr Roigard made a series of statements to the police. He said that on the day of Aaron’s disappearance he had planned to clean up the farm for a property inspection. Part of this entailed moving wood with a tractor and using a wood splitter to break it up. This was a wood splitter set up on the back of the tractor. He said Aaron visited and they got into an argument about the proposed farm purchase. Mr Roigard said that Aaron drove away while he was putting the tractor back into the shed. About 15 minutes later he saw the car at the end of the driveway. He did not see Aaron. He never saw him again.
Much of the Crown case relied on evidence of the imminent exposure of Mr Roigard’s use of Aaron’s savings, evidence of Mr Roigard’s lies about the events of the day, and cellphone evidence, including polling data, which indicated Mr Roigard’s general movements. The Crown said it was a premeditated killing, relying on a number of matters including the evidence of motive and some internet searches Mr Roigard undertook prior to Aaron’s death.[66] What the Crown case lacked was any evidence of how Aaron had died. Without the evidence of Mr F and Mr W, the best the Crown had was evidence of the presence of blood at the Roigard farm as follows:
(a)Six stains of blood on the door of one of the woodsheds. Samples from these stains were consistent with being Mr Roigard’s blood.
(b)A rusty wood splitter located in one of the woodsheds was found to have two tiny “spatter” stains of blood on the handle. No DNA results could be obtained from one of the stains, but a partial profile was obtained from the other, which provided “very strong scientific support” for the proposition it was Aaron’s blood. Ms Knight, a scientist from the Institute of Environmental Science and Research Ltd (ESR), said that spatter stain is produced by a stationary pool or area of blood which has strong force applied to it, although could not exclude other mechanisms.
(c)ESR evidence was that a one metre oval area of ground (comprising soil and woodchips) between the two woodsheds on Mr Roigard’s property gave a positive reaction to luminol when tested. A positive luminol test suggests the presence of blood. This notwithstanding several days of rain had intervened between Aaron’s disappearance and the testing of the site.
On its own, and taken at its highest, this presence of blood was not strong evidence that Aaron had been killed at his parents’ farm. Nor did it provide the mechanism or immediate circumstances of his death.
[66]The Crown relied on evidence that Mr Roigard had done internet research into one-blow killings, and had visited one site that contained material about disposing of bodies.
It was the evidence of Mr F, and to a more limited extent Mr W, that provided the narrative as to what had happened to Aaron. On the Crown case, Aaron and his father had been in the area between the two woodsheds on the day in question. Mr Roigard struck Aaron three times on the back of the head with a wood splitter, killing him. He then scooped up his body with a borrowed tractor, placed it in one of the vehicles at his disposal and drove it to an area where it could be hidden along with the murder weapon.
Treatment of evidence of prison informants at trial
Mr F’s evidence was of reasonably discursive conversations between Mr Roigard and Mr F. But it included claims that Mr Roigard had admitted killing Aaron, described how he did it, and alluded to disposing of his body.
Mr W gave evidence that Mr Roigard made statements consistent with him having killed Aaron and hidden his body. We return in more detail to the content of these statements shortly.
Mr Roigard did not give evidence to contradict these accounts but there is some indication of his position in relation to the truthfulness of their evidence. Mr Keegan, Mr Roigard’s trial counsel, provided an affidavit to the Court of Appeal recording some of Mr Roigard’s instructions to him. The instructions indicated that most of the conversation Mr F described had occurred. As to the conversation with Mr W, there had been questions “asked and answered”, as there would be in a normal interaction between remand prisoners. Mr Roigard’s instructions were, however, that both prison informants were lying when they said that he admitted assaulting or injuring Aaron, or that he knew where Aaron was.
Mr Keegan did not object to the admissibility of this evidence prior to trial. He explained this in his affidavit. He thought it significant that Mr Roigard acknowledged he had spoken to the two witnesses. Mr Keegan believed that the evidence was admissible because, on the “current analysis New Zealand uses”, he believed whether Mr Roigard had made the claimed admissions was a matter for the jury.
At trial, Mr Roigard’s counsel cross-examined both Mr F and Mr W to establish they had been or would be rewarded for their evidence. Mr F and Mr W were also cross-examined on their criminal records to show they had a history of lying. However, counsel did not put to the witnesses that they were lying when they claimed that Mr Roigard confessed to them. Mr Keegan (wrongly) understood the defence to be constrained from doing so if Mr Roigard was not to give evidence.
In closing, Mr Keegan described Mr F as a career criminal, a proven manipulator and liar who had “flowered up” (an expression he took from Mr F’s own evidence) his evidence to falsely claim that Mr Roigard had confessed to him.
In accordance with his instructions, Mr Keegan nevertheless used Mr F’s narrative as a plank of Mr Roigard’s defence. Mr Keegan described the defence in broad brush terms as follows: the Crown had not proved that Aaron was dead, and even if it had, the jury was left with alternatives: was it suicide, murder or manslaughter? As to the last of those alternatives, Mr Keegan pointed to the narrative that, on his argument, Mr F’s evidence tended to suggest that killing Aaron was neither premeditated nor intended.
As to Mr W, Mr Keegan placed him in the same category as Mr F: a man with multiple convictions for dishonesty who was in a lot of trouble, knew that Mr Roigard was charged with a bodiless murder and knew that the police were very interested in getting any information they could. Mr Keegan said that Mr W gave his evidence to get credit on his sentence, and that was what he got.
The trial Judge’s direction
The Judge directed the jury of the need for caution in assessing the evidence of Mr F and Mr W, noting that both had a motive to give false evidence. He told the jury their task was to determine whether that motive affected the reliability of their evidence.
Court of Appeal
In the Court of Appeal, Mr Lithgow for Mr Roigard argued that the Court of Appeal should depart from this Court’s decision in Hudson and proceed on the basis that the evidence of prison informants is presumptively inadmissible as a matter of law.[67]
[67]Roigard v R [2019] NZCA 8 (French, Cooper and Clifford JJ) [CA judgment] at [5], [32] and [36].
Mr Lithgow also advanced an alternative argument – assuming the evidence was properly admissible, the direction the trial Judge gave to the jury in respect of the evidence of Mr F and Mr W was inadequate, having regard to its importance at the trial and the extent to which the evidence had been impeached on cross‑examination.[68] He argued the trial Judge should have emphasised the defence contention that, while Mr Roigard had talked to the two inmates, he had not admitted to killing Aaron. This should have been accompanied by a stronger direction that both Mr F and Mr W had been shown in cross-examination to be utterly unreliable witnesses.[69]
[68]At [69]–[70].
[69]At [72].
Finally, Mr Lithgow argued that trial counsel had erred in running an alternative defence of manslaughter, rather than simply stating that the two witnesses were lying when they gave evidence of a confession.[70]
[70]At [84] and following. There was another ground of appeal, relating to the admissibility of internet search history, which was dismissed by the Court of Appeal: at [105]. Leave to appeal against that finding was declined: Leave judgment, above n 64, at [7].
The issues as argued in the Court of Appeal were therefore quite different to those pursued on this further appeal.
The Court of Appeal rejected the argument that it should depart from this Court’s decision in Hudson because it was binding authority.[71] As to the argument in respect of the Judge’s direction to the jury, the Court of Appeal saw that as a reworking of the first ground of appeal – it was an argument that the Judge should have instructed the jury that the witnesses were not to be believed.[72] It said that while such an approach might be taken:[73]
… where it was clear that the whole of a witness’s evidence was demonstrably false, it would not be appropriate in a case where it was clear that substantial parts of what a witness was saying was in fact correct, and verifiable by reference to known facts.
[71]At [41].
[72]At [73].
[73]At [73].
The Court of Appeal also concluded that trial counsel’s address to the jury was in accordance with Mr Roigard’s instructions so that no miscarriage of justice could have arisen.[74]
Argument on appeal
[74]At [95].
On appeal to this Court, Mr Lithgow argues that the evidence of Mr F and Mr W should have been excluded under s 8 of the Evidence Act 2006. He argues that the application of Hudson has resulted in a presumption in favour of admissibility of prison informant evidence. He argues such an approach is wrong and that Hudson instead requires the court to weigh up the probative value of the evidence against any risk of unfair prejudice that would arise from its admission. In the case of these witnesses, he says Hudson required careful scrutiny of their evidence for the purposes of s 8. That careful scrutiny should have drawn on studies of known miscarriages of justice, and other related studies, which show how easily false secondary confession evidence is fabricated,[75] how hard it is for a defendant to impeach it,[76] and how readily juries overvalue it.[77] Mr Lithgow relies upon the various studies produced by counsel in the appeal in W v R in support of these submissions as to the reliability and unfair prejudice associated with prison informant evidence.
[75]See, for example, Jessica A Roth “Informant Witnesses and the Risk of Wrongful Convictions” (2016) 53 Am Crim L Rev 737 at 780; Russell D Covey “Abolishing Jailhouse Snitch Testimony” (2014) 49 Wake Forest L Rev 1375 at 1381–1382; and Peter P Handy “Jailhouse Informants’ Testimony Gets Scrutiny Commensurate with its Reliability” (2012) 43 McGeorge L Rev 755 at 759.
[76]Covey, above n 75, at 1398–1399 and 1403; and Roth, above n 75, at 780, n 241, citing Sarah M Greathouse “Does Cross‑Examination Help Jurors Detect Deception?” (PhD Dissertation, City University of New York, 2009) which found traditional forms of cross‑examination do not assist jurors in detecting witness deception.
[77]See, for example, Stacy Ann Wetmore, Jeffrey S Neuschatz and Scott D Gronlund “On the power of secondary confession evidence” (2014) 20 Psychology, Crime and Law 339 at 354; and Roth, above n 75, at 773 and 781.
Mr Lithgow submits the two witnesses are correctly described as career criminals, with a history of manipulation and lying, who were effectively paid to give this evidence. Their evidence of confessions is not corroborated by information independent of the police investigation. More than that, Mr F’s evidence was inconsistent with known facts. Overall it had low probative value.
As to the risk that it would have had an unfairly prejudicial effect, Mr Lithgow relies on studies to show that jurors tend to overweight prison informant evidence. In this case, he argues, the calling of such evidence was particularly problematic. The jury would inevitably have formed a very dim view of Mr Roigard and their instinct would have been to convict him of murder. But without the evidence of these two witnesses there was a gap in the Crown case which would have left a reasonable doubt – either about what had happened to Aaron or about whether his death was murder. Against that background, he argues, the jury would want to accept Mr F’s evidence as truthful. No matter how many warning signals there were that it should not be believed, the narrative provided by Mr F would be irresistible because it enabled the jury to fill that evidential gap.
The Crown submits that when undertaking the s 8 analysis, the Court should not displace the role of the jury. It submits that assessing the reliability of the evidence is a matter for the jury. But if the reliability of the evidence is relevant to its probative value for the purposes of s 8, in the case of Mr F, the Crown submits the evidence was sufficiently reliable to be admitted because:
(a)Mr Roigard largely accepted it;
(b)Mr F’s evidence was corroborated by other evidence;
(c)it led police to an item of real significance – a wood splitter which had blood spatter that DNA evidence showed was from Aaron; and
(d)Mr F’s evidence was important to the defence as it supplied the foundation for trial counsel’s closing address on manslaughter.
Framework for analysis
A number of matters are relevant to the assessment of probative value for the purposes of the s 8 assessment. As was held by this Court in W v R, a court may consider not just the significance of the evidence to a matter at issue, but also the reliability of the evidence, when assessing its probative value, before weighing that probative value against the risk of it having an unfairly prejudicial effect.[78]
[78]W v R, above n 63, at [48] and [88(b)] per Glazebrook, O’Regan and Ellen France JJ and [253] per Winkelmann CJ and Williams J.
When undertaking this exercise for incentivised secondary confession evidence, careful scrutiny of the evidence is required.[79] The connection between incentivised secondary confession evidence and miscarriages of justice is well‑established.[80] This evidence is not presumed inadmissible, but nor is it presumed admissible. The careful scrutiny standard means just that: to carefully scrutinise the information provided by the informant, bearing in mind the wider circumstances of both the case and the informant in order to identify and weigh the indicia of reliability that information and those circumstances may contain. The purpose of the scrutiny is to reduce the risk that false accounts of confessions, concocted in the hope or expectation of a reward, will be placed before a jury and so lead to a miscarriage of justice.[81]
[79]We use the term “secondary confession evidence” to mean evidence of statements made by one person alleging another person has admitted guilt – a report by a person to the effect that he or she heard another person (the suspect) confess to a crime: see W v R, above n 63, at [201], n 203 per Winkelmann CJ and Williams J. We also note that beyond the protective threshold of careful scrutiny provided by s 8, there is little by way of a system for checking the reliability of incentivised prison informant evidence. In this respect, we agree with the majority at [55] above that there is a need for additional safeguards such as further guidance for prosecutors and the maintenance of a central register of those who have given evidence as a prison informant and how that evidence was treated. See W v R, above n 63, at [218] per Winkelmann CJ and Williams J.
[80]See the discussion in W v R, above n 63, at [221]–[232] per Winkelmann CJ and Williams J.
[81]At [250]–[251] per Winkelmann CJ and Williams J.
In undertaking that careful scrutiny on this appeal we have had reference to the framework set out in our (minority) reasons in W v R.[82] This framework addresses the particular considerations that apply when secondary confession evidence is offered where criminal justice incentives are in play.[83] It draws upon an analysis of the studies and papers discussed in that judgment – the same studies relied upon by Mr Lithgow.
[82]At [254]–[270] per Winkelmann CJ and Williams J.
[83]We use the term “criminal justice incentives” in the same sense as it is used in W v R, above n 63, at [189] and [201]–[203] per Winkelmann CJ and Williams J.
As that framework makes clear, it is important to ask whether the evidence of a confession has been constructed by the witness to cohere with facts they have gained other than from the claimed confessional statements. It can be relatively easy for skilled liars to construct evidence of plausible sounding confessions in order to obtain criminal justice advantages. The research referred to in Wv R provides striking examples of this occurring in other jurisdictions.[84]
[84]W v R, above n 63, at [241] per Winkelmann CJ and Williams J, referring to 1989-90 Los Angeles County Grand Jury Investigation of the Involvement of Jail House Informants in the Criminal Justice System in Los Angeles County (1990) at 27–31; Peter Cory The Inquiry Regarding Thomas Sophonow: The Investigation, Prosecution and Consideration of Entitlement to Compensation (Manitoba Justice, 2001); Covey above n 75, at 1381; and Roth, above n 75, at 780, n 239. The latter two sources refer to the case of Leslie Vernon White who gave false secondary confession evidence in numerous cases in the 1980s, explaining in 1989 how he had collected information to construct his false evidence, which included calling police and the morgue from jail posing as a police officer or local government official, to obtain non‑public information.
As we come to, the risk of concoction is particularly acute in the case of both witnesses the subject of this appeal. Both had access to a volume of detail about the police case. Both offered evidence in expectation of a criminal justice reward. And both have histories of lying.
The research discussed in W v R also suggests that juries find evidence of prison informants persuasive even when they know it is given in response to incentives.[85] Researchers attribute this to a psychological phenomenon known as the fundamental attribution error – the tendency to attribute the behaviour of others to dispositional factors such as wanting to do the right thing and tell the truth, or to be a good person, rather than to situational factors such as the offering of false evidence in response to an incentive for that evidence.[86]
[85]See the studies referred to in W v R, above n 63, at [233]–[239] per Winkelmann CJ and Williams J.
[86]At [237] per Winkelmann CJ and Williams J, referring to Jeffrey S Neuschatz and others “The Effects of Accomplice Witnesses and Jailhouse Informants on Jury Decision Making” (2008) 32 Law & Hum Behav 137 at 142; Christopher T Robertson and D Alex Winkelman “Incentive, Lies, and Disclosure” (2017) 20 U Pa J Const L 33 at 76; and Evelyn M Maeder and Emily Pica “Secondary Confessions: The Influence (or Lack Thereof) of Incentive Size and Scientific Expert Testimony on Jurors’ Perceptions of Informant Testimony” (2014) 38 Law & Hum Behav 560 at 561, citing Lee Ross “The Intuitive Psychologist and His Shortcomings: Distortions in the Attribution Process” (1977) 10 Advances in Experimental Social Psychology 173 at 174.
The issue of the admissibility of evidence arose pre-trial in W v R. In Mr Roigard’s case, it arises post-conviction. From this perspective we have the advantage, not available pre-trial, of knowing exactly what the witnesses said under oath, including how they responded to defence cross-examination. And we can see the informants’ evidence in the context of the trial as it played out. As this is an appeal against conviction, the ultimate issue is whether, in hindsight, the admission of the evidence has given rise to a miscarriage of justice.[87]
Admissibility of evidence of Mr F
Relevant background
Mr F’s evidence at trial
[87]Criminal Procedure Act 2011, s 232.
In November 2014, Mr F and Mr Roigard were both detained in the same part of Kaitoke Prison. Mr F approached police on 8 November 2014.[88] He offered evidence about conversations he had with Mr Roigard on 1 and 2 November 2014 in the At Risk Unit’s day room. He then spoke to Mr Roigard again on 9 November 2014 and then in late December 2014. He provided written statements to police on 21 November 2014 and 27 March 2015 that provided the basis of his evidence at trial.[89]
[88]This date was in the Crown’s submissions but not in evidence.
[89]These dates also come from the Crown submissions. We did not have copies of these written statements before us.
Mr F claimed that in his initial conversation with Mr Roigard on 1 November 2014, Mr Roigard told him he had been accused of killing his son Aaron, but in his opinion the police did not have much of a case – they did not have a body or a murder weapon. Mr F said Mr Roigard referred to the evidence the police had: cellphone polling data and Mr Roigard’s internet search history, which included a search on one‑punch knockouts and a Mexican Chainsaw Massacre. The latter contained details about disposing of bodies.
Mr F said Mr Roigard told him that the $68,000 Aaron was seeking to get repaid was causing a substantial rift between Mr Roigard and Aaron. Mr Roigard described the police search of the property and said the police had overlooked a wood splitter, which is used for cutting wood, but the police had taken it away at a later stage.
On 2 November 2014, the conversation resumed when the men again were together in the day room. Mr F said that Mr Roigard returned to the police case against him. Initially Mr Roigard said “Oh Aaron will turn up and Aaron will come back”, but this changed as the conversation went on. Mr Roigard later said “Possibly he’d come back” and then “He wouldn’t be back”. Mr F said that over the afternoon Mr Roigard became more comfortable discussing things with him. He said that Aaron had been pressing him for the money and things got quite heated between them, and they had a major disagreement. Aaron had said “Well I’m going to go to the authorities about this” and stormed off. Mr Roigard said he was worried and, as Aaron was walking away, “picked up the splitter and lashed out with it and hit Aaron” three times. Mr Roigard “didn’t mean to do it” – it was “a spur of the moment thing”.
Mr F claimed that Mr Roigard said this had happened where police had found his blood on the farm, but not Aaron’s. On Mr F’s account, Mr Roigard told him that he cleaned up the site with a scoop “or something of that nature” and that he “somehow moved the body [past] the house in either a van or a ute” or a small truck – a vehicle of some kind. Mr F said Mr Roigard “was quite nervous about going past the house because his wife and daughter were both in there” and “that was the scariest part of it, going past the house with his wife and daughter in the house”.
Mr F had a third conversation with Mr Roigard in the day room on 9 November 2014. On this occasion he asked Mr Roigard why he had not told the police it was a “spur of the moment” thing. He claimed Mr Roigard said he “didn’t think they would believe him because of the circumstances and why it had happened and I think about the money came up again”. Mr F said they talked about the beach, because the area of Mr Roigard’s house (Opunake) has a “big beach” and Mr Roigard “indicated that things wash up on the beach” and “he said something about up Eltham Road”. Later in his evidence Mr F changed that to “A long way up Eltham Road” as the actual words Mr Roigard had used and said that this was where Aaron’s body was.
Mr F then saw Mr Roigard either prior to Christmas or in the period between Christmas and New Year. Mr Roigard updated him on the case, that he “hadn’t got bail” and that the police had not found any new evidence. Mr F asked Mr Roigard about what he meant by “Up Eltham Road” and Mr Roigard indicated “something about it being deep and it was on Hastings Road, top of Hastings Road”. Mr F presumed he meant Aaron’s body. Mr F said he initiated the discussion about Aaron’s body because he had some contact with police prior to it and they were interested to “find out”.
Mr F’s criminal history
Mr F has 154 convictions in total, of which 138 are broadly dishonesty-related. Mr F said at trial that most were in relation to a single operation – he was sentenced in February 1997 on 131 charges, of which 122 were dishonesty-related.
Mr F was some years later sentenced to three years’ and one month imprisonment in respect of 15 charges of forgery and related offending. This offending involved the use of documents and was prosecuted by the Serious Fraud Office (SFO). This was a complex fraud, involving Mr F creating some fictional and overstated investment opportunities which he then sold to others with the assistance of documentation he had fraudulently created.
On 30 October 2014, Mr F escaped from custody and confronted and terrorised his estranged wife with a weapon before threatening suicide. His wife escaped and he was arrested and placed in the At Risk Unit at Kaitoke Prison, where he met Mr Roigard. Mr F pleaded guilty to charges including escaping custody and kidnapping.
Benefits received by Mr F
The Crown provided more detail of the benefits received by Mr F for providing statements to police and subsequently giving evidence against Mr Roigard. The Crown confirmed that in submissions filed in support of sentence on the kidnapping offending, the Crown accepted that “a discount in the region of 40% to 60%” was available to Mr F, which included a discount for his guilty plea. At sentence, Mr F received a 50 per cent, two-year sentence reduction. This resulted in a sentence of two years’ imprisonment imposed cumulatively on the existing SFO sentence of three years and one month.
At Mr F’s sentence appeal on the fraud offending, Mr F argued for a further discount on his sentence for his cooperation.[90] This was opposed by the Crown and no discount was given.[91] Finally, after Mr Roigard’s trial, the police, at the request of Mr F’s counsel, wrote a letter for Mr F to use at his subsequent parole hearing, confirming the nature of the assistance he had provided.
Application of framework
Probative value
[90][F] v The Serious Fraud Office [2016] NZHC 271 at [23(b)].
[91]At [57].
The first matter to address under s 8 is the probative value of the evidence. Issues of relevance and reliability bear upon this assessment. As to reliability, on this appeal it is necessary to address the witnesses’ history of lying, the incentives they desired, sought or received, and whether there is any independent corroboration of the claimed confessional statements.
(a) Significance of evidence to matter at issue
The evidence of Mr F was clearly significant to trial issues. If accepted, it established how Aaron died, supported the Crown theory that Mr Roigard’s theft of Aaron’s savings lay behind Aaron’s death, and evidenced Mr Roigard’s responsibility for that death and the subsequent disposal of Aaron’s body. But it is also necessary to address whether the evidence is reliable.
(b) Does Mr F have a record of lying?
Mr F has a history of committing acts of fraud. The offending for which he was in prison at the time he met Mr Roigard involved a complex scheme of fraudulent activity. Mr F created false documents as evidence of a series of investment opportunities which he then sold to investors. He is clearly a person capable of creating plausible false narratives and selling those narratives to prospective investors. To say he is a skilled liar is not an overstatement.
The Crown referred to aspects of Mr F’s evidence which it said had a “ring of truth” about them, such as Mr Roigard’s fear when moving Aaron’s body past the house in which his wife and daughter were. We do not see these details as supporting the truthfulness of the evidence. They are just the sort of detail – easy to concoct but impossible to check – a skilled fraudster would include to sell his story.
It is also relevant that Mr F continued to lie on oath as he gave his evidence against Mr Roigard. Mr F was extensively cross‑examined at trial in relation to his prior offending. He accepted that in the most recent fraudulent offending he had “flowered things up” in connection with the investment opportunities he was selling, but maintained he was offering investment in a legitimate venture with good prospects. He had to accept that this view of his offending was inconsistent with the summary of facts to which he pleaded guilty. He also denied that his offending caused people loss notwithstanding the summary of facts detailing substantial losses.
Mr F was cross‑examined in relation to the offending against his wife. He initially denied that he had kidnapped his wife and tried to characterise the offending as a non-violent, emotional overreaction by him to the end of their marriage. However, the summary of facts to which he pleaded guilty detailed a premeditated and violent incident involving a weapon.
Mr F appealed his sentence for fraud. In cross-examination during Mr Roigard’s trial, Mr F was asked whether he would be bringing up the assistance he gave to authorities in this case in his appeal. He replied unequivocally “No”. When pressed further, he accepted his efforts “may” be brought up but he was unable to “confirm or deny”. In fact, in the High Court a week later, Mr F advanced three substantive grounds of appeal, the second of which was that he was “given insufficient credit for assistance to authorities”.[92] The ground was rejected.
[92][F] v The Serious Fraud Office, above n 90, at [23].
We conclude that Mr F is a person who is experienced at creating complex and plausible false narratives and will go to considerable lengths to do that. He is a liar who is skilled at manipulating others. In fact, given that he mischaracterised the nature and gravity of his previous offending when he gave evidence in this case, in circumstances where this false evidence could be readily checked by the defence, it may be concluded that he is a compulsive liar.
(c) Was the evidence incentivised?
As noted in W v R, the fact that evidence is incentivised is an indication it may be unreliable.[93] In this case, the Crown accepts that Mr F received a reduction in sentence on account of the assistance he provided and also that he received a letter he could use in support of his parole application. But the Crown says there was no agreement reached with him to that effect prior to him offering his evidence.
[93]W v R, above n 63, at [211]–[214] and [257] per Winkelmann CJ and Williams J.
Whether or not agreement was reached, it is clear that Mr F offered his assistance with the expectation of reward, and it is equally clear that he was significantly rewarded. Mr F complained, in the course of his evidence, that he had not received the assistance and support he had expected for having offered his assistance. Although he accepted he received a reduction in sentence for the kidnapping on account of offering to give evidence against Mr Roigard, he said he was unhappy with what he had received: “what, it was indicated to me that I would be, which I would possibly be helped with has not occurred so effectively I am quite unhappy about initially coming here.”
It is also relevant, even if unexplored at trial, that Mr F spent time with police after his first two meetings with Mr Roigard and gained an understanding from the police about what they were interested to find out. He said that on one occasion he initiated discussion with Mr Roigard about where Aaron’s body was because he knew from his contact with the police that that was something they were interested in. When assessing Mr F’s evidence, it is appropriate to weigh the fact that he discussed the case with the police, knew the police were seeking evidence and knew what the police wanted. Mr F offered his statements in confident expectation of reward.
(d) Indications the evidence is unreliable or untrue
The best check on reliability is whether there is independent corroboration of the evidence. The best form of independent corroboration is that the witness’s statement leads to the collection of fresh evidence – such as the discovery of a body. Independent corroboration can also come from the inclusion of detail in the statement of information that could be known only to the offender and could not plausibly come from another source.
The Crown submits there is extensive independent corroboration of the claimed confession. It points to:
(a)Mr Roigard’s admission that he did have conversations with Mr F, in which he told Mr F about the police case against him, just as Mr F claims.
(b)Mr Roigard’s reference to a wood splitter that the police had apparently “overlooked”. The Crown says that as a result of Mr F’s statement, police found fresh evidence – a wood splitter which was found to have Aaron’s blood on it.
(c)Mr Roigard’s admission that the argument and killing took place where Mr Roigard’s blood was found, which fits with the ESR evidence of the presence of blood in an area by the woodshed.
(d)Mr Roigard’s reference to the body being “deep” up Eltham Road/Hastings Road. The Crown says this reference fits, in broad terms, with the phone data about the areas where Mr Roigard’s phone was in use on the day of Aaron’s disappearance.
(e)Mr Roigard’s admission he used a tractor to clean up the site. There was evidence Mr Roigard had a tractor on site, which he had borrowed from a neighbour the day before. It had both bucket and grabber attachments.
Mr Roigard’s admission that he discussed the police case against him with Mr F provides a general corroborative context for Mr F’s evidence – Mr Roigard spoke about the case with Mr F and had the opportunity to make a confession. But these conversations are relevant to the s 8 assessment in another way. The police case, relayed to Mr F by Mr Roigard himself, was a ready source of detail for Mr F to include in his witness statement. As a skilled fraudster, Mr F could use this detail to lend credibility to his statement. There is therefore another narrative that has to be considered – that Mr Roigard told Mr F about the case the police were building against him, but did not confess to the killing.
The need to be cautious in accepting evidence as corroborative is readily apparent in this case. The Crown submits that the evidence about the placement of Aaron’s body is consistent with the cellphone polling data. But on Mr F’s evidence, Mr Roigard discussed that data with him, providing Mr F with detail he could use to shape his evidence.[94] Even so, Mr F could be said to have hedged his bets in the evidence he gave. Mr F gave three possible locations for the body, each of which is at some distance from the other. He claims Mr Roigard said “things wash up on [Opunake] beach”, and that he referred to the body being placed a “long way up Eltham Road” and “on Hastings Road, top of Hastings Road”. If Mr Roigard was confessing to hiding Aaron’s body, why would he name not one, but three different locations kilometres apart?
[94]The Crown put the matter no higher than being consistent with the telephone polling data evidence. That was a proper characterisation. The evidence called at trial on this data was to the effect that there were many variables that could affect which telephone tower a phone polled to. At the end of the evidence, it seemed clear that the data could not show a particular journey by Mr Roigard. It assisted by placing him in a general location.
The Crown relies on Mr F’s account that Mr Roigard used a tractor scoop to clean up the site after killing Aaron. But carefully scrutinised, rather than providing corroboration, this evidence is inconsistent with the known facts. There was evidence that Mr Roigard used his neighbour’s tractor that day, but there was no evidence to suggest the site had been cleaned up with a scoop, which would suggest earth movement. Evidence of scraping or removal of ground cover to hide blood stains would have been obvious to investigators. It was not the Crown case that Mr Roigard used the tractor to clean up the site. Rather, the Crown case was that the tractor was used for the transport and removal of Aaron’s body.
The Crown also points to the blood evidence as corroborating Mr F’s account. It points to the evidence of two tiny spatters of blood on the handle of the wood splitter found in the shed. Mr F said Mr Roigard told him the police had “overlooked” the wood splitter, but later picked it up.[95]
[95]It is possible that Mr Roigard was in fact referring to a mechanical wood splitter in his comments to Mr F. It was not disputed that there was such a splitter at the farm. If that was the case, the statements of Mr F and Mr Roigard could be reconciled, but this issue does not seem to have been clarified in questioning at trial.
As it happens, Mr F was incorrect in one aspect of this evidence. It was Mr F’s statement that led to the wood splitter’s recovery by the police. Mr Lithgow placed a great deal of emphasis on this error as showing that Mr F was lying, but it seems to us an error as to an inconsequential detail – if Mr F was lying, he is unlikely to have been lying about actions involving the police, since such a lie would be easily identified. What is more significant is that when police recovered the wood splitter they found traces of blood on it which, through analysis of a partial DNA profile, ESR attributed to Aaron.
How corroborative of Mr F’s account is the discovery of Aaron’s blood on the splitter? It is true that Mr F could not have obtained knowledge of the significance of the wood splitter from the police case. Police did not identify the presence of the wood splitter or the blood until after Mr F made his statement. His evidence could be said to have led to the discovery of fresh evidence.
The presence of blood on this tool is also consistent with the general narrative advanced by the Crown that the two men were by the woodsheds, when Mr Roigard attacked Aaron, killing him with blows from a wood splitter.
But it is important to remember that this is the narrative provided by Mr F around which the Crown shaped its case. And at trial, the Crown did not allege that the recovered wood splitter was the murder weapon. The Crown could not because the blood evidence was inconsistent with that possibility. The splitter had not been cleaned (a fact established because of the presence of undisturbed rust) yet did not have blood on the striking edge. Rather the Crown case was that the murder weapon had been disposed of. The blood on the recovered splitter was spatters created during the attack on Aaron.
This fact presents a difficulty with Mr F’s evidence on this point. Why would Mr Roigard refer to the police having overlooked a wood splitter when, on the Crown case, the only wood splitter left at the farm was not the murder weapon? Mr Roigard could not have known of the tiny specks of blood on the implement, so why would he have attached significance to an implement that had not been used in the attack?[96]
[96]In his submissions, Mr Lithgow submitted that Mr F had picked up from Mr Roigard that he had been splitting wood with Aaron and then used that fact to construct a narrative. But he said the evidence suggested that the men had been using a mechanical wood splitter rather than manually splitting the wood, and that this was still another flaw in Mr F’s evidence. This point was not fully explored at trial so there is insufficient evidence to form a view on this aspect of the argument.
Overall we agree with the majority that the significance of the evidence of the recovered wood splitter and the presence of Aaron’s blood, on its own, should not be overemphasised.[97] There are other plausible and more mundane explanations for the tiny spots of blood spatter on a farm implement associated with splitting firewood by hand. Aaron was known to help his father with work at the farm. Cuts, abrasions and other injuries are commonplace in such circumstances.
[97]See the reasons of Ellen France J above at [65].
The Crown also points to the corroboration of Mr F’s evidence that Mr Roigard had described how police had found his blood on the door of the woodshed and said that Aaron was killed where that blood was found.
By the time of the conversations between Mr F and Mr Roigard, the presence of Mr Roigard’s blood on that door was a significant part of the police case against Mr Roigard. Mr F said that Mr Roigard told him the police had found his blood. Mr F knew from Mr Roigard that the police attached significance to the presence of Mr Roigard’s blood at the scene. Caution is required.
There are, however, three items of connected evidence relevant to the reliability of Mr F’s claim that Mr Roigard told him he killed Aaron near where Mr Roigard’s blood was found. First, the evidence was that Mr Roigard did in fact injure himself in the hand around the time of Aaron’s disappearance, consistent with Mr Roigard being involved in an attack and injured in the process.[98] Second, there was evidence of Mr Roigard’s blood stains on the woodshed door. Third, and most significantly, luminol testing showed distribution of blood close by, consistent with Mr F’s account of a confession that Aaron was killed where Mr Roigard’s blood was found.
[98]During a walkthrough interview with a police officer at Mr Roigard’s property, Mr Roigard indicated he cut himself and that the cut was “quite deep”. Mrs Roigard also gave evidence that Mr Roigard said he had cut himself and was looking for a plaster, but she also said she did not see any blood.
There was no evidence that Mr Roigard told Mr F about the luminol. Mr F did not mention it in his evidence. If it had been mentioned by Mr Roigard, Mr F would have repeated it either to police or in evidence as he had with other items of evidence Mr Roigard had described to him. Mr F was a voluble and articulate witness with good recall of detail, who readily identified Mr Roigard as the source of that detail.
We consider that in all of these circumstances, these three pieces of evidence, in combination, provide independent corroboration of Mr F’s evidence.
Unfair prejudice
Defence counsel at trial had sufficient material to make the case to the jury that Mr F was an unreliable witness – he was able to cross-examine Mr F about his criminal record, expose Mr F as a liar in connection with that record and put to Mr F the extent of the rewards he received in return for his evidence.
But as discussed in W v R, the principal risk of unfair prejudice associated with incentivised secondary confession evidence is the risk that the jury will overweight the evidence.[99] The reasons for this are traversed in both this judgment and in W v R, but include a tendency on the part of the jury to attribute behaviour to dispositional factors for even the most seasoned criminals giving evidence against a defendant.[100]
[99]W v R, above n 63, at [233]–[239] per Winkelmann CJ and Williams J.
[100]Above at [113]; and W v R, above n 63, at [237] per Winkelmann CJ and Williams J.
In this case, for the reasons advanced by Mr Lithgow, there was such a risk. Mr Roigard was, on any analysis, an unappealing defendant. He had exploited his son in a calculated way, depriving him of his life savings. It was beyond dispute that he had lied about the circumstances of his son’s disappearance. Mr F’s evidence provided the jury with a narrative of what had happened to Aaron. This narrative appeared to be corroborated by objective evidence, at least superficially. We accept that there was an acute risk the jury would fail to adequately weigh all of the indicia of unreliability given that context.
Finally, the Crown submits that when assessing the risk of unfair prejudice it is necessary to weigh that Mr Roigard made use of Mr F’s evidence to support his alternative manslaughter defence. We do not see that as relevant to this assessment. As Mr Keegan made clear, this was a tactical call to deal with Mr F’s evidence, bearing in mind that it was Mr Keegan’s belief that he could not successfully challenge the admissibility of the evidence.[101]
Assessment
[101]See above at [92].
Against this background, and applying the s 8 analysis, we have concluded that Mr F’s evidence was admissible. Its probative value outweighed the risk of an unfairly prejudicial effect.
It is true this was evidence of a confession offered by a skilled fraudster in confident expectation that he would be rewarded for it. It is also true that Mr F is a proven liar who lied in the witness box. But the evidence of the presence of Mr Roigard’s blood on the shed door, and of blood on the ground nearby, provides some independent corroboration of the narrative that Aaron was killed in the general area of the sheds.
The risk of unfair prejudice remained, but that was addressed by the Judge in the directions he gave in connection with Mr F’s evidence, as set out above.[102]
Admissibility of evidence of Mr W
Relevant background
Mr W’s evidence at trial
[102]See above at [97].
Mr W gave evidence that in late 2015 he was remanded in custody at Kaitoke Prison in a cell close to Mr Roigard’s cell. This was shortly before the trial against Mr Roigard began.
On a day when he was in the shower area at the same time as Mr Roigard, he overheard Mr Roigard telling another inmate that the police had been trying to get someone to spy on him while he was in prison to collect information. He said that after that occasion he spoke to Mr Roigard a few times and that he found it strange that Mr Roigard always wanted to talk about his case.
At one point in his evidence, Mr W claimed Mr Roigard told him that the police would “never find his son”, “never find the body”, and that Aaron “got what he deserved”. Mr W said that Mr Roigard told him about the case the police were assembling against him, including that they were looking for Aaron’s body around the Eltham Road area. But later in his evidence, Mr W claimed that Mr Roigard said Aaron had gone to Australia. In cross-examination, he confirmed that Mr Roigard said that when he got out of prison he intended to look for his son.
On Mr W’s account, Mr Roigard told him that he and Aaron had argued at the house because Aaron’s partner or one of his friends had found out that “the farm thing wasn’t true … and yeah, it was all backfiring”. After their argument, they went for a drive in Aaron’s car. When they came back, Mr Roigard had left the car parked at the end of the driveway. Mr Roigard did not say where Aaron was when Mr Roigard parked the car there.
Mr Roigard told Mr W that someone had said “he attacked [Aaron] on the front lawn but then moved him with the tractor and he said the tractor wouldn’t fit on the lawn … so they must be lying”. Mr Roigard also talked to Mr W about Aaron’s money. When asked whether Mr Roigard had spent it, Mr W replied “Nah, well, I don’t really like believe half of what he said anyway, it was just … you could never tell whether he was just fantasising or telling the truth”.
Mr W gave evidence that suggested Mr Roigard had discussed the police case against him in detail, including the evidence of cellphone data. He gave his evidence in a halting way, mumbling, and claiming that his memory was “not the flashest”.
Mr W’s criminal history
At the time of his conversations with Mr Roigard, Mr W was on remand on charges of dealing methamphetamine, receiving and unlawful possession of firearms. Mr W has 112 prior convictions. He has 64 convictions for dishonesty-related offending, including a conviction for giving false details to police as to his identity (he gave false details when stopped by police when he was driving without a licence) and for making a false statement or declaration. The cross-examination of Mr W focused principally on these prior convictions.
Benefits received by Mr W
At his sentencing on 24 November 2015, Mr W received a discount of 22 months from a starting point of three years and six months’ imprisonment. His sentence of 20 months’ imprisonment was commuted to 10 months’ home detention. However, the Crown notes this discount included credit for assistance in an unrelated matter, which resulted in the police recovering high-powered weapons.
Crown reliance on Mr W’s evidence
In both opening and closing, the Crown placed some reliance on Mr W’s evidence. In closing, the Crown acknowledged that he received a discount on his sentence. However, it sought to address this by pointing out, first, that this was credit for assistance in both the prosecution of Mr Roigard and the unrelated recovery of high‑powered firearms and, secondly, by reminding the jury that when Mr W was asked why he made the statement to the police, he replied it was not for credit but rather “because I believed he murdered his son”.[103]
Application of framework
Probative value
(a) Significance of evidence to matter at issue
[103]The Crown also reminded the jury that Mr W’s belief was not relevant to their decision‑making but rather to their assessment of Mr W’s credibility.
The evidence was clearly significant. The comment “he got what he deserved” could be read as an admission that Mr Roigard had killed Aaron. Relevant also is Mr W’s claim that Mr Roigard said he and Aaron argued about the money on the day and that they went for a drive in Aaron’s car. This latter point was inconsistent with Mr Roigard’s statements to the police. He claimed that the two had stayed on the farm until Aaron left and that it was Aaron who had parked the car at the end of the driveway.
(b) Does Mr W have a record of lying?
Mr W has 64 convictions for dishonesty. He could not be described as a sophisticated fraudster but he has shown a willingness to lie for his own advantage.
(c) Was the evidence incentivised?
The evidence was incentivised.[104] Mr W received a substantial discount for the assistance he gave. Mr W also received reward for other assistance he provided to police, which tends to show that he was significantly motivated at that time to obtain reduction in the time he would serve and had an understanding of the system by which he would do that.
(d) Indications the evidence is unreliable or untrue
[104]Above at [169].
The description of the car trip is corroborated by the evidence of Jack Scott, a local farmer’s son, who said he saw Aaron in his car that day with another person. Otherwise Mr W’s evidence generally coheres with the case that police had assembled against Mr Roigard. But this does not provide independent corroboration of Mr W’s evidence. On Mr W’s own evidence, the police case, as recounted to him by Mr Roigard, is a plausible source of these otherwise seemingly corroborating details. Mr W was clear in his evidence that the police case against Mr Roigard was their principal topic of conversation. It is also relevant that Mr W gave contradictory accounts – on the one hand, that Mr Roigard said Aaron might have gone to Australia but, on the other hand, that Aaron “got what he deserved”.
Unfair prejudice
The risk of unfair prejudice is the same as with Mr F: the risk that the jury would have failed to adequately weigh the indicia of unreliability associated with this evidence and this witness.
Assessment
Mr W was an incentivised witness with a history of dishonesty. The evidence he gave was lacking in detail and there was a ready source for the scant detail it did contain – Mr W’s conversations with Mr Roigard in which Mr Roigard told Mr W of the police case against him. There was no independent corroboration of the evidence to meet these concerns. In our view, this evidence was unreliable. As to unfair prejudice, the evidence carried with it the risk that the jury would overweight the evidence, discounting the significance of the incentives and Mr W’s criminal history in favour of the explanation offered as to why Mr W came forward to give his evidence. Overall, we conclude that the probative value of this evidence is outweighed by the risk of unfair prejudice attaching to it. The evidence should have been excluded.
Section 30 of the Evidence Act
An alternative argument was advanced for Mr Roigard – that the evidence was unfairly obtained for the purposes of s 30 of the Evidence Act as it was “bought” evidence. We agree with the majority that, at its heart, this argument rests on a challenge to this Court’s decision in Hudson.[105] Leave to appeal was granted on the basis that the decision in Hudson would not be revisited.[106]
Conclusion
[105]See the reasons of Ellen France J above at [78].
[106]Leave judgment, above n 64, at [3].
We consider the evidence of Mr F was admissible. The evidence of Mr W should have been excluded under s 8 of the Evidence Act on the basis its probative value was outweighed by the risk of an unfairly prejudicial effect.
Outcome on appeal
A finding that evidence was wrongly admitted does not automatically lead to a successful appeal. The ultimate issue on this appeal is whether the wrongful admission of this evidence has given rise to a miscarriage of justice in the sense its admission created a real risk that the outcome of the trial was affected, or resulted in an unfair trial.[107] Since a majority of this Court is of the opinion that the appeal is to be dismissed, we do not address this issue.
[107]Criminal Procedure Act, s 232; and see Misa v R [2019] NZSC 134.
Solicitors:
Crown Law Office, Wellington for Respondent
at [76]–[85].
4
6
0