|
|
Home
| Databases
| WorldLII
| Search
| Feedback
Court of Appeal of Fiji |
COURT OF APPEAL OF FIJI
LALAGAVESI
v
STATE
Court of Appeal
Shameem, Goundar and Scutt JJA
11 February, 7 March 2008
Criminal evidence - Identification - Visual identification - Inherent dangers - Turnbull principle - Circumstances of identification - Recognition - Accused charged with sacrilege - No one identifying accused as having broken into temple - Witnesses chasing intruder from outside adjacent house - Police eye witness stating accused well known to him - Accused convicted in magistrates' court - Appeal to High Court dismissed - Magistrate failing to give herself Turnbull direction on dangers of convicting on visual identification evidence - Whether such failure rendering conviction unsafe - Whether improper inferences of fact drawn - Whether appeal to Court of Appeal to be allowed.
Early one morning PW1 observed a man (`one Fijian man wearing a pompon hat on his head, a T-shirt and ¾ pants') endeavouring to remove shutters from a neighbour's house. PW1 and two others chased the intruder from a distance before losing him. Upon returning home, PW1 discovered the door to a temple in his compound was open, statues were displaced and $8 was missing. The appellant, L, was charged with sacrilege. No identification parade was held and L denied that he was the person responsible for the offence. PW1 identified L from a photograph at the police station out of about 50 police photographs. He identified L in court, saying he recognised his face from seeing him near the end of the chase. PW2, a police officer who had joined the chase, said that he had seen L face to face during that chase and that L was 'well known to him'. The magistrate relied on the identification evidence of PW2 and found the circumstantial evidence against L to be strong: he had been seen only a few metres from the temple, at about 6am, attempting to break in to the neighbour's house. The magistrate considered that the possibility that someone else broke into the temple was so remote as to be fanciful. L was convicted and appealed unsuccessfully to the High Court. However, neither the magistrates' court nor the High Court had adverted to the principles in R v Turnbull [1976] 3 All ER 549, approved by the Court of Appeal in State v Wainiqolo [2006] FJCA 70, concerning the proper approach and care that should be taken in relation to visual identification evidence. L, who was not legally represented in his original trial nor in his appeal to the High Court, appealed with leave to the Court of Appeal, which had to consider whether the magistrate should have cautioned herself in accordance with Turnbull on the quality of the identification evidence against the appellant. The state conceded that the appeal should succeed.
HELD: Appeal allowed.
In the instant case, the need to 'examine closely the circumstances in which the identification by each witness came to be made' was a crucial requirement for consideration which had been overlooked in both the magistrates' court and in the High Court and which explicitly raised the questions listed in R v Turnbull [1976] 3 All 13R 549. On the evidence before the court, no one had identified L breaking in to the temple or running therefrom. The temple could have been broken into at any time from the previous afternoon, well before the 'Fijian man' was sighted attempting the housebreaking and L was chased, until the time of discovery. The only identification as to the attempted housebreaking was by PW1, who identified a man wearing a T-shirt. Later, after the chase, PW1's identification was of a man wearing a 'black jacket' (as per the evidence of PW2)-not a T-shirt. The identification of L by PW2 was not in relation to the attempted housebreaking, which he did not see (nor of the temple break-in, of which he knew nothing at that time), but to the chase in which he (PW2) engaged together with PW1. In her ruling, the magistrate had effectively accepted that PW2's knowledge of Mr Lalagavesi as a 'known offender' and PW2's identifying a person in the neighbourhood as a known offender, namely L, was sufficient to draw the inference that L was both the attempted housebreaker and the party who committed the temple sacrilege. Had the magistrate's mind adverted to the Turnbull principles, the danger inherent in that pattern of thinking might have been, and should have been, exposed.
The identification was not at the scene of the crime nor contiguous with it, but at two locations removed from it. That L was a known offender, combined with his 'running', led to the conclusion that L was (first) the attempted housebreaker, then (secondly) the desecrator of the temple. In those circumstances, the need for the magistrate explicitly to warn and guide herself in accordance with Turnbull became crucial. The dangers inherent in identifying a known offender, then inferring that such individual 'must' be the person involved in the crime, were not inconsequential. Turnbull guarded against the dangers of identification away from the scene of the crime and relating that identification back to the crime, because the person identified was already known-particularly as a known offender. L's trial was rendered unfair, and his conviction unsafe, by such danger not being addressed by the magistrate in accordance with Turnbull and Wainiqolo. The magistrate had erred in law. In upholding the magistrate's decision, through not addressing the failure by the magistrate to warn herself of the danger of relying entirely upon the `identification' evidence of PW2 to convict L, the High Court had also erred in law. The Turnbull warning should have been given by the trial magistrate. Its oversight and omission in the magistrates' court should have led inexorably to L's High Court appeal having been allowed. The conviction and sentence were set aside and the unsatisfactory nature of the evidence precluded any retrial (see paras [I], [21]--[22], [27], [37]-[39], [43]-[44], [71], [86]-[87], [93], [100]-[101], [111]-[121], below). Dicta of Lord Widgery CJ in R v Turnbull [1976] 3 All BR 549 at 551-552 and State v Wainiqolo [2006] FJCA 70 applied.
Cases referred to in judgment
R v Turnbull [1976] 3 All ER 549, UK CA
State v Wainiqolo [2006] FJCA 70, Fiji CA
Tekuru v State [2004] FJHC 94, Fiji HC
Legislation referred to in judgment
Court of Appeal Act (Cap 12), s 22(1)
Penal Code, s 298
Appeal
The appellant, Saula Lalagavesi, appealed against the decision of the High Court dismissing his appeal against conviction in the magistrates' court of sacrilege contrary to s 298 of the Penal Code (Cap 17). The facts are set out in the judgment of the court.
V Vosarogo for the appellant.
P Bulamainaivalu for the state.
7 March 2008. The following judgment of the court was delivered.
SHAMEEM, GOUNDAR and SCUTT JJA.
INTRODUCTION
[1] The determination of this court is that the appellant, Mr Lalagavesi, should succeed in his appeal. This accords with the principles relating to reliance upon identification evidence, as set out in R v Turnbull [1976] 3 All ER 549, approved by the Court of Appeal in State v Wainiqolo [2006] FJCA 70.
[2] In convicting Mr Lalagavesi, the magistrates' court did not advert to Turnbull. In his appeal against conviction, again the principle in Turnbull was not adverted to by the High Court.
[3] The appeal goes to a particular aspect of identification evidence directly involving the circumstances relating to the identification, the use to which it can be put and the care that must be taken by trial courts in ensuring that Turnbull and Wainiqolo are applied with rigour. This is of fundamental importance to the trial process. The need to pinpoint the error with clarity is essential. Hence, we set out in some detail the history, evidence in the trial and the magistrates' court determination, and the issues raised in this appeal by Mr Lalagavesi, his counsel and counsel for the state.
GROUNDS OF APPEAL
[4] The appellant, Mr Saula Lalagavesi, appeals against his conviction on 8 February 2007 by the magistrates' court at Lautoka. On that day Mr Lalagavesi was convicted under s 298 of the Penal Code (Cap 17) and sentenced to two years' imprisonment. He appealed to the High Court against conviction. On 4 April 2007 the High Court dismissed his appeal.
[5] By letter dated 26 April 2007 Mr Lalagavesi lodged an application in the Court of Appeal seeking leave to appeal in respect of both conviction and sentence. On 23 August 2007 he was granted leave to appeal to the Full Court of this court.
[6] Mr Lalagavesi was not represented in his original trial, nor in his appeal or leave to appeal application. He lodged his own appeal papers to the Full Court. However, he was represented in the hearing before this court. Mr Lalagavesi's lack of legal representation was raised by him, by his counsel in the Full Court of the Court of Appeal, and by counsel for the state as a matter of concern, particularly upon the basis that had he been represented at the trial, the requirement for a Turnbull direction would have been raised.
[7] In his appeal to the Full Court of the Court of Appeal, Mr Lalagavesi is entitled to appeal not on the facts, but on law only: see Court of Appeal Act (Cap 12), s 22(1).
[8] In his leave to appeal application, the Court of Appeal observed that Mr Lalagavesi's –
'only argument ... was that he had consistently denied he had committed the offence and complained particularly that no identification parade was held.'
The question for the Full Court as set out in the ruling on leave to appeal is:
'Whether the learned magistrate should have cautioned herself in accordance with Turnbull's case on the quality of the identification evidence against the appellant: R v Turnbull [1976] 3 All ER 549 (CA).'
APPELLANT'S AND RESPONDENT'S SUBMISSIONS
[9] As initially unrepresented, Mr Lalagavesi made his own written submissions to the Full Court, focusing on the identity question. Amongst other matters he said that the issue of identity-
'was important... , and it was not established that [he] was the person who was in the actual crime ... Reasonable doubt exists in the allegation because the only evidence leading to my conviction was the evidence of identification which was not fairly conducted. There are contradictions in the identification of the accused by witnesses and the description of clothes worn by the accused ... Finally, I ... submit that my identity was not established and the state fails the identification criteria with regards to the Turnbull case to the "line of sight", "direct vision" [and] "identity parade" [in] which the state witness[es] ... failed miserably, it is submitted that discrepancies in identity [are] to be treated with care as a slight mistaken identity will result in the conviction of an innocent person or accused ...'
[10] The state agrees that Mr Lalagavesi should succeed in his appeal, saying amongst other matters:
'Having perused the learned magistrate's and judge's decisions, it is apparent that the former had failed to caution herself of the danger of relying entirely on the uncorroborated identification evidence of [the arresting police office] PC 3018 ... [PW2] to convict the appellant.'
[11] Counsel for Mr Lalagavesi observes that the appeal is from a decision where the conviction was based 'entirely on identification evidence', going on to acknowledge the state's agreement that 'conviction was based solely on the identification evidence of one of the prosecution witnesses' and adding:
'Noting the approval of the Court of Appeal in State v Wainiqolo [2006] FJCA 70 on the guidelines [as to identification evidence] in Turnbull the three critical issues are: (i) Did the magistrate warn herself of the danger of conviction on identification evidence alone? (ii) Did the magistrate direct her mind on the circumstances on which the identification came to be made? (iii) Did the magistrate remind herself of any specific weakness which had appeared in the identification evidence?'
C MAGISTRATE'S JUDGMENT
[12] In her judgment, the magistrate observed that on the afternoon of 11 October 2006 PW1 closed and bolted the door of the small temple in his compound. Awoken by a noise early in the morning of 12 October, from his window PW1 saw some six metres distant a Fijian man dressed in a jacket and pompon hat endeavouring to remove shutters from a neighbour's house. (PW1's actual evidence was that he saw 'a man ... trying to remove the shutters from my neighbour's house ... He removed the shutters and then he removed the window ...')
[13] Alerting a neighbour, PW1 and two others chased the intruder from a distance before losing him. Upon returning home, PW1 discovered the door to the temple was open, statues were displaced and $8 was missing.
[14] There was no dispute as to the breaking and entering of the temple or the theft of $8. What was in dispute was the allegation that Mr Lalagavesi was the person responsible.
[15] Of the three people who chased the intruder, one was PC 3018 (PW2). He said Mr Lalagavesi was 'well known to him' for Mr Lalagavesi had attended at the police station many times for different cases and had spent three months there as an extramural prisoner, with PW2 seeing and greeting him. PW2 said he had recognised Mr Lalagavesi during the chase. He recorded his own statement later on that day, 12 October 2006.
[16] PW1 identified Mr Lalagavesi in a photograph at the police station out of about 50 police photographs. He identified Mr Lalagavesi in court, saying he had seen him clearly at two points in the chase, and that Mr Lalagavesi's face 'was familiar' as he had 'seen him around the area'. The magistrate was critical in some respects of this evidence and the conduct of the police vis-à-vis photographic identification rather than an identification parade. Setting no great weight on the identification evidence of PWl, the magistrate said:
'The use of photographs is permissible only if it is not possible to hold [an identification] parade for some reason. If the identification is to be done by way of photographs then the officer who organises this should give evidence so the court can be sure proper procedures were followed.'
[17] It was insufficient, the magistrate said, for the prosecution to rely on a question to Mr Lalagavesi 'Do you want to do identification parade?' and the recorded answer 'No' [which Mr Lalagavesi disputed] 'to establish that the accused refused to participate in a parade'. [18] The magistrate nonetheless concluded:
'In this case I am satisfied that [PW1] did chase the accused for some distance and that he had time and opportunity to see at least the general stature and appearance of the accused. His evidence is therefore supportive of the evidence of [PW2] whilst not being conclusive.'
[19] Mr Lalagavesi's evidence was that on the night of 11-12 October 2006 he was staying at his brother's house at Fultala Place Field 40. His brother's evidence was that Mr Lalagavesi had indeed spent the night there, as he said. However, Mr Lalagavesi's brother said that after going to bed at 9pm on 11 October he did not see Mr Lalagavesi, and that upon rising and going off to work at 5.45am on 12 October, he did not see him. Mr Lalagavesi said he left his brother's house at 6am to return to his father's house at Captain Withers Street Field 40. This was consistent with his record of interview, except that when asked if he went anywhere early in the morning of 12 October, Mr Lalagavesi answered 'No'. In oral evidence he denied knowing, or even having ever seen, PW2.
[20] The magistrate concluded:
'Fultala Place and Captain Withers Street are very close together and not far away from Chandmari Street where part of the chase took place. [Mr Lalagavesi] admits he was out alone in the Captain Withers/Fultala area at 6am. The prosecution cases rests almost entirely on the identification of [PW2]. This officer has 6 years' experience. He says he got a close look at the accused's face on two occasions during the chase. The accused is well known to him, and he identified him by name immediately. I reject the evidence of [Mr Lalagavesi] that he did not know and had never met [PW2]. I am satisfied beyond reasonable doubt that [Mr Lalagavesi] has been correctly identified as the intruder who was attempting to break into the neighbour's house.'
[21] This court has a number of reservations in respect of this conclusion, to which we will return in the context of Turnbull. Save as to say at this point that what is compelling is not that the identification relates to Mr Lalagavesi at the attempted housebreaking (as incorrectly stated by the magistrate), but that it relates to Mr Lalagavesi as 'chasee': that is, the identification of Mr Lalagavesi by PW2 was not in relation to the attempted housebreaking which he did not see (nor of the temple break-in, of which he knew nothing at that time), but to the chase in which he (PW2) engaged together with PW 1.
[22] On this, see later, however, at this point it is important to observe that no identification by PW2 relates to the time of the attempted housebreaking, much less the temple. The only identification as to the attempted housebreaking is by PW1. That identification is not of Mr Lalagavesi-at least as Mr Lalagavesi-but of 'one Fijian man wearing a pompon hat on his head, a T-shirt and ¾ pants'. Later, at the field after the chase, PW1's identification was of a man wearing a 'black jacket' (as per the evidence of PW2)-not a T-shirt.
[23] As to the attempted break-in of the neighbour's house, even had Mr Lalagavesi's identification by PW2 occurred in relation to it, would this mean that the identification can then be related to the desecration of the temple?
[24] The magistrate dealt with this by reference to the principles governing reliance upon circumstantial evidence, citing State v Wainiqolo [2006] FJCA 70 and Tekuru v State [2004] FJHC 94.
[25] The magistrate referred to the following statement of principle from Wainiqolo:
'It appears to be a widely held misapprehension that circumstantial evidence is not proper evidence and that it cannot prove the guilt of an accused. That is not correct. If the circumstantial evidence is relevant to the allegation, it will be admissible. In general it is unlikely to be as telling as direct evidence but, in some cases, it may be more persuasive as, for example, where there are a number of matters of circumstantial evidence from different sources which all point to the same conclusion.'
[26] From Tekuru the magistrate cited this passage:
'The law is that where the prosecution relies on circumstantial evidence to prove its case, the accused can be found guilty only if the only reasonable inference the court can draw from the set of circumstances, is the guilt of the accused. There must be no other available and reasonable inference which is equally consistent with the accused's innocence, and the court must be satisfied of the accused's guilt beyond reasonable doubt.'
[27] This court doubts that the evidence here meets the standards as recited in Wainiqolo and Tekuru, albeit the magistrate said she adopted them. We return to this later in canvassing the evidence and application of Turnbull.
[28] In the magistrates' court circumstantial evidence was dealt with as follows:
'The circumstantial evidence in this case is strong. [Mr Lalagavesi] was seen only a few metres from the temple at about 6am. He was attempting to break in to the neighbour's house.'
[29] We interpose here to observe again that Mr Lalagavesi qua Mr Lalagavesi was not identified at this point. As earlier noted, the evidence of PW1, the only person who saw the attempted break-in of the house, was as to 'one Fijian man ...' See further later.
[30] The magistrates' court continued as to circumstantial evidence:
'The temple had been shut up only the afternoon before and is close to the complainant's house. I consider that the possibility that someone else broke into the temple is so remote that it could be described as fanciful.'
[31] Yet this recitation of the circumstances contains a non-sequiter: proximity to PW's house is not so much an issue as proximity to the neighbour's house. Identification is relevant to proximity to PW1's house-for the purpose of establishing PW1's ability to see clearly or otherwise, so as to identify or fail to identify the attempted housebreaker. However, as to the break-in of the temple, it was not PW1's house upon which the attempted housebreaking occurred or was seen, but the neighbour's house. Hence more relevant to whether the temple was broken into by the person who attempted the housebreaking is whether the temple was proximate to the neighbour's (not PW1's house).
[32] Further on circumstantial evidence, the magistrate continued:
'The accused raised a number of issues in his defence. It is true that the prosecution did not find clothing of the description worn by the intruder. It is true that the $8 was not recovered and that police were unable to get any clear fingerprints from inside the temple. The prosecution does not have an obligation to present every possible piece of evidence to the court. It does bear the burden of adducing evidence sufficient to establish that the accused committed the offence charged beyond reasonable doubt. I am satisfied that the prosecution have discharged this burden and you are convicted as charged.'
IDENTIFICATION EVIDENCE AND TURNBULL
[33] In written submissions, counsel for Mr Lalagavesi observed that upon perusal of the magistrates' court record, proposition 3.8 (i)-(i) Did the magistrate warn herself of the danger of conviction on identification evidence alone?-‘was never considered by the magistrate and not considered on appeal to the High Court’.
[34] As to the second proposition-(ii) Did the magistrate direct her mind on the circumstance on which the identification came to be made?-the written submissions say:
'It is also clear that some discussion of the evidence on (ii) was made ...'
[35] Counsel's reference here is to the magistrate's statement:
'I consider that the possibility that someone else broke into the temple is so remote that it could be described as fanciful.'
[36] In regard to the third proposition-(iii) Did the magistrate remind herself of any specific weakness which had appeared in the identification evidence?-the written submissions say that the magistrate 'failed entirely to discuss anything to remind herself [as] would be required by [proposition] (iii).
[37] Written submissions by counsel for the state set out the principle in these terms:
'As a matter of practice, the uncorroborated identification evidence of ... [PW2] meant that it was mandatory of the learned magistrate to have warned herself of the danger of relying entirely on such evidence to convict the appellant. In the absence of such caution, the learned magistrate did err in law and the learned judge in upholding the former's decision ... committed the same error.'
[38] The passages from Turnbull [1976] 3 All FR 549 at 551-552 per Lord Widgery CJ, upon which both counsel rely, and to which Mr Lalagavesi in his written submissions alluded, say:
'First, whenever the case against an accused depends wholly or substantially on the correctness of one or more identifications of the accused which the defence alleges to be mistaken, the judge should warn the jury of the special need for caution before convicting the accused in reliance on the correctness of the identification or identifications. In addition [the judge] should instruct them as to the reason for the need for such a warning and should make some reference to the possibility that a mistaken witness can be a convincing one and that a number of such witnesses can all be mistaken. Provided this is done in clear terms the judge need not use any particular form of words. Secondly the judge should direct the jury to examine closely the circumstances in which the identification by each witness came to be made. How long did the witness have the accused under observation? At what distance? In what light? Was the observation impeded in any way, as for example by passing traffic or a press of people? Had the witness ever seen the accused before? How often? If only occasionally, had he any special reason for remembering the accused? How long elapsed between the original observation and the subsequent identification to the police? Are there any material discrepancies between the description of the accused given to the police by the witness when first seen by them and his actual appearance? If in any case, whether it is being dealt with summarily or on indictment, the prosecution have reason to believe that there is such a material discrepancy they should supply the accused or his legal advisers with particulars of the description the police were first given. In all cases if the accused asks to be given particulars of such descriptions, the prosecution should supply them. Finally, [the judge] should remind the jury of any specific weaknesses which had appeared in the identification evidence.'
[39] For this court, the need to 'examine closely the circumstances in which the identification by each witness came to be made' is a key focus requiring consideration in Mr Lalagavesi's case. It was overlooked in both the magistrates' court and in the High Court appeal.
[40] The 'circumstances in which the identification by each witness came to be made' are crucial, raising explicitly the questions listed in Turnbull, namely had the witness ever seen the accused before, how often, had he any special reason for remembering the accused ...
[41] In Turnbull these questions were posed in the particular circumstances existing there. In the present case, the court must look at the particular circumstances existing vis-à-vis the sighting by PW1 at the neighbour's house-then the sighting and identification by PW1 and more particularly PW2 not at the neighbour's house but extraneous to it-at a shop and in a compound 'near the fence' (see further later). Crucially, this all occurs in a context where the identifier PW2 knows Mr Lalagavesi as a `known offender'.
TRIAL EVIDENCE OF IDENTIFICATION
[42] A fundamental problem in this case is the elision occurring, from the outset, in the prosecution's case vis-à-vis identification.
[43] As observed earlier, PW2 did not see the attempted break-in of the neighbour's house. PW1 alone saw this. As noted, his identification was not relied upon by the magistrate, whose reliance was placed upon the identification evidence of PW2. Yet neither PW2 nor PW1 saw any break-in to the temple when it occurred. Nor (apart from the time PW1 locked it) has either of them any real knowledge as to when it happened.
[44] The temple break-in came to attention when PW1 returned to his home after the chase. On the evidence before the court, the only time that can be fixed upon as pinpointing the temple break-in is that given by PW1- that he locked up the temple on the afternoon of 11 October 2006. Hence, it could have been broken into at any time after the afternoon of 11 October 2006, well before the 'Fijian man' was sighted attempting the housebreaking and well before Mr Lalagavesi was chased. The only 'timing' of the break-in is reliant upon the inference drawn by the magistrate that it took place before the attempted housebreaking. She did not consider the possibility that it took place well before the attempted housebreaking, or after the attempted housebreaking-whether by the 'Fijian man with pompon hat'-who (if he were a person other than Mr Lalagavesi and this possibility should have at least been considered) could have hidden whilst PWl, PW2 and the other man (PW2's brother, RN) went off on the chase leading to the identification of Mr Lalagavesi as the person being chased. Alternatively, if the 'Fijian man with pompon hat' were Mr Lalagavesi, then the possibility must exist that some person other than Mr Lalagavesi took advantage of the absence of PW1, the temple owner, PW2, the police officer, and RN, his brother, from the compound and broke in.
[45] The magistrate dismissed these possibilities not by turning her mind to them or giving them consideration but by ruling any alternative out as 'fanciful'. This assessment was made without adversion to what any such possibility might be. It was also made without the magistrate's giving herself a Turnbull warning.
[46] The continual putting of Mr Lalagavesi at the scene of the temple break-in, despite his never having been identified there, nor even identified as Mr Lalagavesi at the housebreaking, is characteristic of this case from the outset. At the bail application on 20 October 2006 the magistrates' court notes record the prosecution as saying:
'Strongly object bail. Very common in Fiji for temples to be broken into and accused not caught. Accused was chased by police officers from scene and caught...' (Our emphasis.)
[47] Mr Lalagavesi was never 'chased from the scene'-whether `the scene' was the temple (unknown at that time to have been broken into and possibly not having been broken into at that time) or the neighbour's house.
[48] The chase began not 'at the scene' but 'across the road', commencing:
[49] The trial took place on 1 February 2007, with three witnesses called by the prosecution: PW1, PW2 and PW3 (a police officer who became involved on or about 19 October 2006 but who, in any event, was not involved in the chase, nor was `at the scene' on the day of the offence-namely 16 October 2006). PW2's brother RN did not give evidence.
[50] In evidence-in-chief, identification evidence provide by PW1 was as follows:
'... 12/ 10/2006 I was asleep in my room. I heard noise and I woke up. I saw from my window one Fijian man wearing a pompon hat on his head, a T-shirt and ¾ pants. The man was trying to remove the shutters from my neighbour's house. Six meters away. It is a separate house. This man is the accused [in dock]. He removed the shutters and then he removed the window. I opened my house door then I called my neighbour and the man saw me and ran away. I saw my neighbour and the neighbour saw the man too.'
[51] We interpose at this point to observe that it is not PW2's evidence ('my neighbour', `the neighbour') that he saw the man at this time. PW2's evidence is that he saw a man running across the road and commenced chasing him. Further, PW1's evidence here conflicts with PW1's own evidence in cross-examination, when PW1 asserts that he `saw your face clearly two or three times' `when I was sitting on my bed'.
[52] PW1's evidence continues:
'Then three of us chased him. [PW2] and [his] brother-[RN] and I was the third. We chased accused to Chan Mani Road, then he tried to cross a private compound and then he ran back to Chan Mani Road, then to Oriana Place, he went in the drain and one of the police officers followed him, ie [PW2]. I and [PW2] blocked the other end and we surrounded accused. Accused went into a compound, jumped the fence and ran away then he disappeared. [PW2] said he identified accused when he was inside the fence we saw his face. I came home and saw all the things inside the temple were misplaced.'
[53] We again interpose to note further inconsistency in PW1's evidence at this point. In regard to the following of Mr Lalagavesi into the drain, and the blocking of the other end and surrounding of Mr Lalagavesi, PW1 refers to PW2 both times-that is, according to PW1's evidence, PW2 is the police officer who followed the man being chased, into the drain: `he went in the drain and one of the police officers followed him, i e [PW2]'. PW1 then says `I and PW2' 'blocked the other end and ... surrounded the accused'. Can PW2 be both 'following him (the party chased)' and 'blocking the other end'?
[54] On the photographic identification of Mr Lalagavesi, in cross-examination PWl says:
'Police showed me about 50 photos. I recognized his face from seeing him inside the fence. I just identified accused by the photographs.'
[55] We note that PW1 does not say he 'recognized' Mr Lalagavesi from (say) the chase across the road, at the Chand Mani shop, or-more importantly-from `seeing him' (as 'one Fijian man') attempting the housebreak. The association between Mr Lalagavesi's photograph and Mr Lalagavesi seen in person is in respect of seeing Mr Lalagavesi in the field-'inside the fence'. Identification at this point is not connected to a crime of attempted crime, but to a chase.
[56] PWl then gives an 'identification' he has not given in evidence-in-chief and which appears to be entirely new:
'I saw my neighbour walking around and when you saw him you sat down facing me?
How did you recognise me so far away and wearing a pompon? When I was sitting on my bed, I saw your face clearly two or three times.'
[57] This was not given in evidence-in-chief and there is no previous or other reference to PW1's having seen Mr Lalagavesi or identified him at this point-that is, while PW1 was 'sitting on his bed'-when (it is understood) he saw 'one Fijian man'. Never before has he said he `saw [this person's] face clearly two or three times', much less that it was Mr Lalagavesi's face he saw. This may be the or a reason for the magistrate's having decided not to rely upon PW1's evidence of identification.
[58] A similar disjunction appears in the `identification' evidence of PW2, as given in evidence-in-chief and in cross-examination (see later).
[59] Continuing in cross-examination, PWl then reverts to his original evidence in response to Mr Lalagavesi's question:
'How can you identify me in a photo without a pompon. When we were running after you and you were inside the fence we saw you face clearly.' (Our emphasis.)
[60] PW2's evidence of identification began with his confirmation that his neighbours include PW1 and others, and continues:
'12/10/06. I was at home in the morning. While I was sleeping, my brother woke me up and said someone is at the neighbour's house doing a break-in. I woke up and came outside and by then the culprit had left the house.' (Our emphasis.)
[61] That is, PW2 never identified 'one Fijian man' at the house, much less the temple, and much less Mr Lalagavesi. Relevantly at this point PW2 acknowledges the lapse in time taken to wake up and come outside. Should this raise a question in the magistrates' court's mind as to why, in light of the delay and the disappearance of the person sighted attempting the break-in, if Mr Lalagavesi were the `one Fijian man' he is still to be seen 'running across the road' rather than having disappeared entirely? Being detected in an act of criminality or attempted criminality should surely render the culprit fleet of foot rather than leaving them to be sighted minutes later 'running across the road'? Mr Lalagavesi lived only some ten minutes away. How many minutes did it take for PW1 to raise his neighbour (RN), for RN to wake PW2, for PW2 to get up, come out, confer-'he was there', 'he went that way', 'he must have turned down ... 'gone around ...'then start looking for the suspect?
[62] PW2 continues his evidence-in-chief:
'I then came on the main road and saw a Fijian boy wearing black jacket, pompon and ¾ shorts running across the road. i followed the boy and gave chase. We went to shop at Chand Mani, then we were face to face. He then turned back and ran to Field 40. At Field 40 Junction we were again face to face and he was surrounded by some other people including [KS]. Then accused jumped into a compound at Field 40 and I followed him. He picked up some empty beer bottles and he jumped the other side. I followed him across the road and then I lost track of him. I identified him-first time and second time we were face to face.' (Our emphasis)
[63] PW2's evidence crystallises:
'I know his name, Saula Lalagavesi. I know him because he used to come to Police Station for many cases, also he did Extramural Prisoner at Police Station. I talked to him at Police Station-used to greet each other when we met. After that I then came and reported to Police and I told them the name of the suspect. Accused is present in court. Identifies accused in dock.'
[64] In cross-examination, amongst other questions and answers, on the matter of identification PW2 answered as follows:
'What time were you woken?
Cc. 5.45am, not sure, because I didn't take a watch. In your statement you say 6am?
I will recall exact time.
Your brother [RN] who woke you up?
[RN] ...
Did you see me break into the temple?
No ...
Was an identification parade conducted?
No, because I know you very well. And I will be giving evidence in court. Investigation Officer asked accused regarding identification parade, but I said I would be identifying him and he said that's ok...
12/10/06-your statement.
Doesn't mention temple?
I only knew [J's] house was broken into. Later I found out temple was broken into.
Fijian man with pompon-how do you know it was me?
I know accused very well and we were twice face to face. 1 saw the black spot on the left side of his face.
[Court; Note accused has a black spot/mole on his left cheek.]'
[65] Here, we note-as we consider the magistrate ought to have noted at the trial-that the mole (or birth mark-see Mr Lalagavesi's evidence) is identified when PW2 is `face to face' with Mr Lalagavesi-never at any time directly associated with the break-in at the temple or the attempted break-in of the neighbour's house. Yet again, the elision occurs-as if identification of Mr Lalagavesi through 'his mole' puts him at the scene of the crime or attempted crime, or somehow supports an inference that he was there.
[66] The cross-examination evidence from PW2 continues
'You didn't say "black spot" in your statement?
If I had not known his name and known him well, then I would have given an accurate description but I knew him well...
Did you see me break into the temple?
No ...
Why did you say I did it? You were running away and I chased you ... Twice I came face to face with you. One time I went around the house and came face to face with you and the second time when you had to cross the drain at Field 40 ...
What time?
I can't say time. It was daylight, after 6 when we gave chase ...'
[67] It is here that PW2 makes a statement as to identification that has never appeared in evidence previously: `One time I went around the house and came face to face with you ...' Compounding this is the immediately following clause `and the second time when you had to cross the drain at Field 40 ...'
[68] 'True it is that PW2's evidence previously has been that the second time he saw Mr Lalagavesi 'face to face' was at the field. However, all PW2’s evidence up to now has been that the first time he saw Mr Lalagavesi 'face to face' in the chase was at the Chand Mani shop-never 'around the house'.
[69] As to time, the statement: 'it was daylight, after 6 when we gave chase' provides the court with three different times given by PW2-6am in his statement, 5.45am in cross-examination; then later in cross-examination, 'after when we gave chase'.
[70] In re-examination PW2 said be 'identified accused when I was the courtroom's width away from him'.
[71] Identification of Mr Lalagavesi at that distance may well have been sufficiently close to identify him as Mr Lalagavesi without quibble. Yet where does this conclusion take the identification issue? What does that identification mean? The aim of the trial is not to prove that Mr Lalagavesi was 'in the field' or 'at the Chand Mani shops'. It is not even to determine whether Mr Lalagavesi attempted the break-in of the neighbour's house (albeit these matters can be relevant as circumstantial evidence-but see our earlier concerns on this aspect). The aim is to determine whether Mr Lalagavesi broke in to PW1's temple.
[72] Not only did PW2 not see anyone-including Mr Lalagavesi-break into the temple or attempt to break-in to the neighbours house, the identification comes in the course of a chase-not at the temple, nor at the house (or 'around the house') but commencing on the main road:
'I then came on the main road and saw a Fijian boy wearing black jacket, pompon and 3/a shorts running across the road.'
[73] Contrary to PW2's statement in cross-examination, in his evidence-in-chief there is no reference to his having seen anyone upon 'going around the house'. There is no reference to his 'coming face to face' with anyone upon 'going around the house'-Mr Lalagavesi or anyone else, even, one Fijian boy'.
[74] Rather, after having said he came on the main road to see 'a Fijian boy ... running across the road', as earlier noted, PW2 nominated two occasions only upon which he says he identified Mr Lalagavesi:
'I followed the boy and gave chase. We went to shop at Chand Mani, then we were face to face. He then turned back and ran to Field 40. At Field 40 Junction we were we were again face to face and he was surrounded by some other people including [KS]. Then accused jumped into a compound ... I identified him-first time and second time we were face to face.' (Our emphasis.)
[75] In re-examination PW2 said that he 'identified accused when I was the courtroom's width away from him'.
[76] PW3's evidence was that he went to the scene of the temple break in and also recorded the statement of PW1. He said that PW2 'told me he saw accused and he chased him. He particularly identified accused, ..'
[77] PW3 said that he identified the accused on 19 October 2006. That is, PW3's identification is not connected with the neighbour's house, nor the temple, nor indeed even with the chase. It relates solely to the identification of Mr Lalagavesi as Mr Lalagavesi-a person with a record, a 'known offender' and the person identified by PW2 at a shop and in a field, neither location being temple or neighbour's house, and by PW1 in the field and in a photograph at the police station.
[78] Mr Lalagavesi elected to give sworn evidence in which he denied the offence.
[79] In cross-examination it was said to him that PW2 'identified you at [the] scene. Why would he lie? ...'
[80] This question lacks foundation entirely. It ought not to have been put. Were Mr Lalagavesi represented at trial, objection would or should have been taken. PW2 did not 'identify' Mr Lalagavesi 'at the scene'. As with earlier evidence, an elision occurs: the idea that identification has taken place 'at the scene' of the temple break-in, because it has taken place elsewhere; or that the identification at the shop and in the field is the same as or constitutes identification at the scene of the break-in. This is not so. This is not even an improper inference. It is an assertion made without a basis. It is unsustainable, highlighting the need to take into account the 'circumstances' of identification as cautioned in Turnbull.
[81] The submissions made by the prosecution confirm yet again the need for caution:
'Accused charged with sacrilege. Temple was broken into $8 cash taken. Straight after that house next door attempted break-in, [chased] by [PW1] and PW2 across three streets position identified by a police officer who knows accused very well through extra-mural prisoner at police station. Three months at Police Station. Accused stays just 10 minutes walk to crime scene.'
[82] As noted earlier, why, then, if Mr Lalagavesi was indeed the person attempting the break-in at the neighbour's house, was he still 'running across the road'-after the time it must have taken for PW1 to alert his neighbour PW2 and PW2's brother RN, for PW2 to waken, dress, get up, come outside, confer and then go out of the compound to the road?
[83] The link is made between the chase and the attempted break-in, as if they are precisely contiguous: the chase without question being of the person attempting the break-in, from the point of the attempted break-in.
[84] Further, the prosecution's submission that the temple was broken before the attempted housebreaking is no more than assumption. As earlier observed, the temple was secured the afternoon before-without oversight until PW1 returned from the chase. Fair enough it is to make the inference, but that it is inference must be stated, its basis spelled out. Blunt assertion is misleading.
[85] The prosecution's submissions continue:
'Early morning he says he was home with his brother. Brother says he did not see him that morning. Very clear with case. Police proved case beyond reasonable doubt.'
[86] If anything were proved beyond reasonable doubt, it was that:
[87] All the above must raise in the eyes of a court the possibility that Mr Lalagavesi's identification:
is not only not based upon direct evidence, but cannot be based upon inferences properly drawn. Rather, the identification rests upon an inference wrongly drawn, namely that as Mr Lalagavesi is a known offender and he was seen (first identification as Mr Lalagavesi) at a nearby shop, running when chased by three men including a police officer, he must have been the offender.
[88] As to the running, even then, when Mr Lalagavesi `runs', he stops in the field to pick up two beer bottles-presumably for their recovery or exchange value. If a person who is a `known offender' finds himself being chased by three men, including a police officer, his running may be understood, whether having committed an offence or not. Yet had he committed an offence is it likely he would, whilst running to escape capture vis-à-vis that offence, pause to collect beer bottles-particularly when alleged to have had in his possession $8 stolen from the temple? There was no suggestion at any time by the prosecution or prosecution witnesses that Mr Lalagavesi was picking up beer bottles for a defensive or offensive ploy. He was never questioned to this effect and the magistrate made no such comment as indeed no comment of this nature could or should have been made.
[89] Mr Lalagavesi's being identified as a `known offender' is intimately linked with:
1. the inference drawn by PW2 that he was the person who should be chased as having (so PW2 concluded) inevitably being the person who attempted the neighbourhood house break-in (at this stage, the chasers having no knowledge or suspicion of any temple break-in: they concluded they were chasing an attempted house burglar) and
2. once the temple break-in had been discovered, the inference drawn by PW2 that he was the person responsible.
[90] in her ruling, the magistrate effectively accepted that:
was sufficient upon which to draw the inference that Mr Lalagavesi was both the attempted housebreaker and the party who committed the temple sacrilege.
[91] Had her mind adverted to the Turnbull principle, the danger inherent in this pattern of thinking may have been, and should have been, exposed.
TRIAL EVIDENCE AND TURNBULL
[92] In Wainiqolo the Court of Appeal described the judge's direction to the assessors as 'careful and extensive', given 'clearly [with the] guidelines in [Turnbull] in mind'.
[93] The judge's direction in Wainigolo bears repeating in full:
'This is a case where the State relies upon the accuracy of an identification of the accused and the defence contends she is mistaken. Where that is so, I should warn you of the special need for care before relying on the identification evidence alone as the basis for a conviction.
The reason for that is that experience has shown that it is quite possible for a perfectly honest witness to be mistaken about identification. An honest witness who is convinced of the accuracy of what he or she says may well come across as convincing but may still be mistaken. Bear in mind that we all make mistakes in thinking that we recognise people, even those we know well. That is not to say that you cannot rely on identification evidence. Of course you may, but you need to be careful in deciding whether the evidence is good enough to be relied upon. Can I suggest that you think about the circumstances under which this witness saw the accused at the time in question. How long did the witness have the accused under observation? What sort of distance were they away from each other? What was the lighting like? Had the identification witness ever seen the accused before? Did the identification witness know the accused and, if so, how well? Was there anything about the situation which would cause the identification witness to take particular note? Think about these sorts of issues carefully to see if you can rely upon the evidence of identification given by the first prosecution witness.' (Our emphasis.)
[94] The Court of Appeal dismissed the appeal in Wainiqolo, both in regard to conviction and to sentence. Insofar as identification evidence was in issue, the court found wholly against the accused's grounds of appeal. The facts in Wainiqolo are important, and worth recounting in the context of the present case, for they are so disparate from what the magistrates' court had before it in Mr Lalagavesi's trial. The distinction is important.
[95] As set out by the trial judge in the context of Turnbull, the facts in Wainiqolo were:
'The [witness] can only have caught a quick glimpse of the robber who took off his balaclava in the process of searching for something inside the vehicle. They were clearly quite close to each other; at the most according to [the cousin] some one metre away. It was a bright morning so there was good lighting but her view of the robber would have been impeded by the car structures, the roof, the door jamb, etc. She was in fear. The witness was, it seems, a distant relative of the accused. They grew up in the same village. She knew him by sight but not really to speak with. It will be a matter for you to determine just how well she knew the accused prior to this incident. Take into account that prior to the incident she says she hadn't seen him for a long time but did recently catch sight of him sometime in December 2004 or January 2005 ... Against these matters you must weigh the warning that I earlier gave you about the possibility of identification witnesses being perfectly honest and convinced of the accuracy but nonetheless mistaken.'
[96] The Court of Appeal saw this recitation as a `very fair and proper application of the Turnbull guidelines [with] no ground for criticism'.
[97] In Wainiqolo the identification witness was driving out of her home to bank some $13,000 takings for the New Year period from the nightclub she owned with her husband. Having left the compound, she was waiting for her teenage cousin to close the gate when a vehicle was driven up close to her car, preventing it from being driven further. Four masked men, armed with an axe and cane knives, ran out and up to her car. The car windscreen was smashed with the axe, and one of the men attempted to reach the car keys. In so doing, he removed his balaclava so that the woman was able to see his face. Recognising him, she said: 'Semisi, you can't do this to me. I know you.' After replying he didn't know her, he threw the keys back into the car. Rejoining the others, he was heard by her to say: 'Someone has recognized me'.
[98] That is, the identification was 'on the spot'. It was identification on the scene of the crime. It was an identification where the witness saw the accused face to face in the close vicinity of her car, which was the immediate target of the attack: the takings were under the driver's seat. The purpose of the attack was to rob the driver by removing the $13,000 takings from the car and hence from her.
[99] In the present case, the identification is remote from the scene of the crime and the crime is discovered after the identification has taken place. The danger in the present case is that identification of a person who is known to the identifier as a person with a criminal record, and well known at that, when the person's attachment to the crime is only that they were sighted 'running across the road' after another crime had been attempted in a compound proximate to the road. Indeed, their sighting 'running across the road' is not the identification sighting.
[100] Rather the first identification sighting occurs even more remotely from the attempted crime of housebreaking-not the crime the subject of the charge. It occurs at a shop away from the compound, away from the house, and away from the temple (Chand Mani). The second identification sighting is even more remote from the first-in a field further away from the scene of what becomes known as the crime after the chase and the identification sightings.
[101] The task then is to return to the questions in Turnbull, and thence to address Lord Widgery's statement as to the importance of the 'circumstances in which the identification by each witness came to be made'. All the while, in the addressing of these questions, the circumstances must be central: namely, that the identification is not at the scene of the crime, nor contiguous with it. It is at two locations removed from the scene of the attempted crime, and the scene of the later discovered crime with which the person identified is then a charged.
[102] Turnbull: how long did the witness have the accused under observation? PW2 had the accused under observation on two occasions upon which he identified him as Mr Lalagavesi-at the Chand Mani shop and on the compound by the fence.
[103] Turnbull: at what distance? On each occasion, face to face.
[104] Turnbull. in what light? According to PW2, in the early morning light at some time shortly after about 5.45am, or shortly after 6am, or sometime after Earn.
[105] Turnbull: was the observation impeded in any way, as for example by passing traffic or a press of people? Insofar as the identification at Chand Mani shop, there is no indication of whether or not people were around; as to identification in the field, there were others-people including PW2, PW1 and PW's brother RN. J, the person whose house PW1 observed as being attempted to be broken into does not appear to have participated in the chase.
[106] Turnbull: had the witness ever seen the accused before? Evidence of both PW1 and PW2-Yes.
[107] Turnbull: how often?
PW2-
'On many occasions at the police station in the context of his being an offender generally and in the extra mural programme.' PW1 (in cross-examination)
'Why did you tell police in your statement that you knew me? I had seen you a lot of times before that day.
Where?
Field 40 Junction on the roadside. He was new at Field 40 at the time. How can you know me, when I don't know you and I haven't seen you? I stay there for fifteen years and I saw you when you came, as a new person to that place ...
You don't know me?
I know you. I have seen you a lot of times. You only knew me because [PW2] told you? No'
PW 1 (in re-examination)-
'I know accused by face. Sometimes I saw him walking along the road. I don't know him personally.'
[108] Turnbull: if only occasionally, had he any special reason for remembering the accused? Albeit on the evidence of neither was it 'occasionally', at least for PW2 there was a 'special reason for remembering' Mr Lalagavesi-that PW2 had seen him at the police station generally and in relation to the extra mural programme-that is, as a known offender.
[109] Turnbull, how long elapsed between the original observation and the subsequent identification to the police? For PW2, this was immediate, as PW2 was the police. Further, PW2 reported to the police after the chase. For PW1, it appears this occurred much later after the event (12 October 2006)-as it appears a week later:
'You didn't come to the police station to identify me?
No, I did come to police station and I did identify you. They showed me a lot of photos and I pointed out from the photos. A lot of photos were in the police file. Police showed me about 50 photos. I recognised his face from seeing him inside the fence. I just identified accused by the photographs. I can't recall when but maybe one week later.
Why can't you remember when you looked at the photos?
I can't remember the date. I know his face. I can remember his face.' (Our emphasis.)
[110] Turnbull: are there any material discrepancies between the description of the accused given to the police by the witnesses when first seen by them and his actual appearance? There was dispute about the clothing of the upper body. PW1 identified a 'T-shirt' initially, whilst PW2 identified a 'black jacket'. Later, PW1 changed 'T shirt' to 'black jacket'.
[111] Turnbull: are there any specific weaknesses appearing in the identification evidence? The most pronounced problem here lies in the key issue: the danger of identifying someone being a known offender, and linking this person, by reason of that knowledge, back to an offence (and, in particular, an offence not even known to have been committed at the time of 'identification'). On PW2's evidence, there can be little doubt that the person he identified at the Chand Mani shop was Mr Lalagavesi, nor that the person he identified in the field was Mr Lalagavesi. Yet how far does that take the court in the identification of Mr Lalagavesi as the person who was at the scene of the attempted housebreak-or as the person engaging in the attempted housebreak? How far does it take the court in the identification of Mr Lalagavesi as the person who broke into the temple and stole $8?
[112] Turnbull: The circumstances in which the identification by each witness came to be made are crucial. Unlike Wainiqolo, this was not a case of 'on the spot' or 'at the scene of the crime' identification. It was a case of identification 'off the spot' or 'away from the scene of the crime'-some streets or distance away, at least-and identification of a person because the person was a known offender and that he was running. That he was a known offender, combined with his 'running', leads to the conclusion that Mr Lalagavesi is (first) the attempted housebreaker, then (secondly) the deseci'ator of the temple. In these circumstances, the need for the magistrate explicitly to warn and guide herself in accordance with Turnbull and Wainiqolo becomes crucial.
[113] Criminal investigations are often warned to avoid fixing upon one line of investigation, at least too early, to the detriment of considering other possibilities. The dangers inherent in identifying a known offender, then inferring that that individual 'must' be the person involved in the instant crime, are not inconsequential. Turnbull guards not only against the dangers of identifying and being certain about identification where an individual is seen right at the scene of the crime. It guards against identification away from the scene of the crime and relating that identification back to the crime, because the person identified is already known-particularly as a known offender.
[114] Mr Lalagavesi suffered from this danger. His trial was rendered unfair, his conviction unsafe, by this danger and its not being addressed by the magistrate in accordance with Turnbull and Wainiqolo. In this oversight, the magistrate erred in law.
[115] In upholding the magistrate's decision, through not addressing the failure by the magistrate to warn herself of the danger of relying entirely upon the 'identification' evidence of PW2 to convict Mr Lalagavesi, the High Court erred in law.
DETERMINATION
[116] The appeal is an appeal on a question of law. The Turnbull warning should have been given by the trial magistrate. Its oversight and omission in the magistrates' court should have led inexorably to Mr Lalagavesi's High Court appeal having been allowed.
[117] In its absence, this court is satisfied that the conviction is unsafe and unsatisfactory. A miscarriage of justice has occurred.
[118] The conviction and sentence must be set aside. The unsatisfactory nature of the evidence means there should be no retrial.
ORDERS
[119] The appeal is allowed.
[120] The conviction and sentence of the appellant determined by the magistrates' court at Lautoka on 8 February 2007 are quashed.
[121] There shall be no retrial of the appellant.
Solicitors:
Legal Aid Commission (Suva) for the appellant.
Off ice of the Director of Public Prosecutions (Suva) for the state.
PacLII:
Copyright Policy
|
Disclaimers
|
Privacy Policy
|
Feedback
URL: http://www.paclii.org/fj/cases/FJCA/2008/118.html
