Irish Court of Criminal Appeal
You are here: BAILII >> Databases >> Irish Court of Criminal Appeal >> Director of Public Prosecutions -v- Crowe [2015] IECCA 9 (03 February 2015)
URL: http://www.bailii.org/ie/cases/IECCA/2015/C9.html
Cite as: [2015] IECCA 9
[New search] [Help]
Judgment
___________________________________________________________________________ | ||||||||||||||||||||
THE COURT OF APPEAL Record No: 112/2012 Ryan J. Sheehan J. Edwards J. The People at the Suit of the Director of Public Prosecutions Respondent V
Christopher Crowe Appellant Judgment of the Court delivered on the 3rd of February 2015 by Mr. Justice Edwards. Introduction. 2. The case was concerned with the sending by telephone of a message which was menacing to a member of An Garda Siochána, Detective Sergeant Denis Smith on the 27th of November 2008. 3. Following his conviction, the appellant was sentenced, on the 19th of December 2011, to imprisonment for a term of three years for the offence in question, backdated to the 8th of December 2011, i.e., to the date on which he went into custody. 4. By way of additional relevant background it should be stated that on the same date the appellant was also sentenced by the same court in respect of an offence of endangerment, to which offence he had pleaded guilty. The endangerment offence had been committed while he was on bail in respect of the offence of sending a menacing message, which necessitated the imposition upon him of a consecutive sentence for that offence. The learned trial judge therefore imposed a further three year sentence in respect of the endangerment offence consecutive to the three year sentence imposed for the offence of sending a menacing message, but suspended the final twelve months of that sentence. 5. The appellant appeals against both his conviction and sentence in respect of the offence of sending a menacing message. Evidence before the jury 7. The witness noted that the incoming call was from a private number. He answered it and it was immediately clear that it was a male caller. The caller asked him "Is that Mr Smith?" and the witness said, "Yes." The caller then said: "Robert Smith?" and at the same time as he was saying "Robert Smith" the witness said, "Denis Smith" by way of clarification. The caller then said: "Damien Smith, or anyway it's you that I want" and he then said: "You were up on a cross pointing down at me. I want €10,000 for your murder". There was a pause for a moment and then the caller added: "You're a dead man" and, "I want you now". “Understand”. “You’re up on a cross”. 8. The call terminated at that stage. It had all happened very quickly and the witness, whose initial reaction was shock, claimed that he did not have time to speak other than his own name. The jury later heard evidence in relation to telephone records which established that the duration of the call was 32 seconds. 9. Detective Sergeant Smith interpreted the reference to him being up on a cross as a possible reference to him being viewed through a telescopic sight, and he was alarmed. 10. He then telephoned his immediate superior in An Garda Siochána, Detective Inspector Peter O’Boyle, and reported the incident. Later that morning the witness also recounted what had occurred to Superintendent John Quirke, who was the officer in charge of his district. Detective Inspector O’Boyle ordered an immediate investigation. As Detective Sergeant Smith was the injured party, he was deliberately distanced from the investigation. 11. Detective Sergeant Smith was stood down as a witness at this point but returned to the witness box later in the case to give the following further evidence. He told the jury that on the following day, the 28th of November 2008, he went to Clondalkin Garda station. At Clondalkin Garda station he was shown a video recording of an interview that had taken place on the previous day between the Gardaí and the appellant. He stated that he immediately recognised the voice of the appellant as the voice that had spoken to him on the previous day on the telephone. He agreed in cross examination that at half seven in the morning, if a person had just got up, that person might be “a little croaky”, particularly if “he had been up boozing all night or smoking all night”. He further accepted that such a person wouldn’t necessarily sound like they would sound in an interview room. 12. The witness stated that he had listened to the recording in Clondalkin Garda Station for approximately 15 seconds, before indicating his recognition of the voice. 13. Detective Sergeant Smith agreed that he knew the appellant before, and that the appellant would have known him. However, when he saw the appellant being interviewed on the video tape, and heard that he had been arrested on suspicion of having made the phonecall, he had wondered what it was all about, as he had had no difficulty with the appellant and the appellant had nothing to do with any case that he was working on. 14. The witness agreed under cross examination that before he heard the voice on the tape he knew the circumstances of the appellant’s arrest, and the details of a mobile phone that had been recovered at the time of arrest. However, he added: “But that wasn’t what I based the recognition on. The recognition was on viewing the video and listening to the voice” “And it immediately jumped out at me, it is the same voice, it’s the same person.” Asked if he was 100% certain, he stated: “I am 100% certain that the voices are the same”. 15. Detective Sergeant Smith was pressed in cross-examination to identify what it was that made him so certain, and he replied “There’s nothing I can put my finger on.” Further pressed in respect of this, he accepted that the same words used during the threat were not spoken on the recording, that there was nothing distinctive in the timbre of the voice recording, and that he could identify no particular accent in the voice recording. He described the accent as nondescript. The voice possibly exhibited a Dublin accent but it was not pronounced. He stated: “There is nothing specific in his accent that would have made me think otherwise, think other than that it was him.” When it was put to the witness that his capacity to identify the voice was nothing more than his gut instinct, he responded:
Q. You would have done? A. I did.” 17. The Jury also heard evidence that at 20.30 on the 27th of November 2008 Detective Garda Eamon O’Neill in the company of colleagues attended at Room 111 at Bewley’s Hotel, Newland’s Cross, Dublin 22. These Gardaí had in their possession a warrant to search it. There were four persons present in the room, a female (Miss Kevanne McNamara, partner of the appellant), a 7 year old male child (the son of Miss McNamara and the appellant), the appellant himself, and another male, (a Mr. John Byrne). 18. Upon the Gardaí entering into the room it was observed that the appellant was asleep, and that located on a pillow to the right hand side of the appellant’s head, was a mobile telephone which was seized. The appellant, upon being woken, accepted that it was his phone. The number of this phone was given in evidence before the jury. The keypad was locked with a security code. It was subsequently established that this mobile telephone was the mobile telephone upon which the call was made to Sergeant Smith at 07.12 that morning. The appellant was arrested at 20.35 p.m. on suspicion that he had committed an offence contrary to Section 5 of the Non Fatal Offences Against the Person Act 1997, and he was then taken to Clondalkin Garda Station where he was detained pursuant to the provisions of Section 4 of the Criminal Justice Act 1984. 19. The appellant was interviewed on three occasions while in custody for the purposes of advancing the investigation. 20. The first such interview was between 22.26 and 23.30 on the 27th of November 2008. In the course of that interview the appellant again admitted that the phone that had been found on his pillow was his. He stated that he had owned it for just two weeks, and he readily provided Gardaí with the PIN number for it. He admitted that the phone also had a security code number, which was known only to him, and he stated that he locked the keypad with that code most of the time. He refused to make the security code number available to Gardaí claiming that there was private material on the phone from his girlfriend. He claimed that he had gone to sleep at half past twelve on the previous day and had slept until woken by the Gardaí. He was unable to say when John Byrne had arrived in the room. He denied making a call to Detective Sergeant Denis Smith at 07.12, but accepted that only a person with the security code number could have made such a call. He denied knowing Denis Smith and stated that the only person he knew in Ballyfermot was Robert Smith. He was further asked:
"Answer: I didn't use the phone. Can Garda Smith identify my voice? Surely he can identify the person who rang him. Why would I be threatening to kill him? "Question: It was a male who rang him, not your girlfriend or the child? "Answer: Well, it wasn't me. 22. The third and final interview was between 17.26 and 18.02 on the 28th of November 2008. The appellant again denied making the phonecall. Asked to say who had done so, he responded: “Haven't a clue. I'll find out when I get out who did it because the only person could have been John Byrne.” It was put to him that John Byrne had not arrived at the hotel until 5pm on the previous day (27th), and he initially said he didn’t know, and then said “If you said he wasn’t there til 5 he didn’t”. 23. Kevanne McNamara and John Byrne were both called as witnesses for the defence. Kevanne McNamara told the jury that the call had been made by John Byrne, while the appellant was asleep. She stated that she thought it was only “for a mess”, and that they didn’t realise it was a Garda. She stated that she had provided John Byrne with the security code number. She accepted in cross-examination that she had lied to the Gardaí in making a statement denying any knowledge of the call and asserting specifically that John Byrne had not made the call. Her explanation was “I didn’t realise what was going on”. She was robustly and forensically cross-examined but could offer no further explanation. She said that she told John Byrne that he would have to tell the truth. However, she could not explain why she herself did not come forward long before the trial, when it was manifest that John had taken no steps in that regard, and seek to correct her earlier statement. She denied that she had concocted a story for the trial to try to help her partner and that she had given false testimony to the jury. 24. John Byrne, in turn, told the jury that he was a friend of the appellant, and that on the evening before the call was made he went up to the hotel where the appellant and his girlfriend were staying to have a drink with them. He claimed that he was in and out of the hotel room many times on that evening. After the child went to bed he was having a few drinks and remained drinking in the room all night. He became drunk and said that early the following morning: “I had a bright idea that hour of the morning to make a few prank phone calls.” His own mobile phone was out of credit. The appellant was asleep. He picked up the appellant’s phone but couldn’t get the security lock off. He said Kevanne provided him with this. He rang one or two of his friends and then a random number in the contacts stored on the phone. It turned out to be Detective Sergeant Smith’s number. He claimed he didn’t know Denis Smith or that he was a Garda. 25. Mr Byrne was also robustly and forensically cross-examined. Under cross-examination he claimed to be unable to recount exactly what he may have said to the persons who answered his prank calls. He was unable to explain the reference to Robert Smith. It was put to him that in a statement he had made to the Gardaí he had told them he didn’t arrive in the hotel room until after 5pm on the 27th of November 2008 because he had been watching a Champions League game on TV. He accepted that he had said that and could offer no explanation other than to claim he had been confused. He also could offer no explanation for not coming forward sooner. He also denied concocting a story for the trial, or that he had given false testimony to the jury. 26. The appellant was convicted by unanimous verdict of the jury after a total deliberation time of 2 hours and 49 minutes. Grounds of Appeal Against Conviction
2. The learned trial Judge erred in law and in fact or in a mixture of law and fact in admitting evidence as to the capacity of the Injured Party, Detective Sergeant Denis Smith to identify with any degree of certainty the voice of the Appellant herein with reference to having heard a portion of approximately 15 seconds of the Appellant's voice when he was being interviewed having been arrested for the subject matter of the within Indictment. 3. The learned sentencing Judge in law or in fact or in a mixture of law or fact in refusing an application for a direction that the Appellant herein had no case to answer upon the conclusion of the prosecution. 4. The learned trial Judge erred in law or in fact in admitting the evidence proffered by the Prosecution in circumstances where the manner in which the purported voice recognition took place, did not safe guard or protect the rights of the Accused man. 5. The learned trial Judge erred in law or in fact or in a mixture of law or fact in admitting for the purposes of voice recognition, such portion of the Defendant's interviews as were conducted by the prosecuting and investigating members, in circumstances where the Defendant had been given the ordinary caution with reference to anything that he may say may be taken down in writing and given in evidence against him. At no stage was there a specific warning given that he may incriminate himself by simply speaking words. 28. The learned trial judge had conducted a voir dire in relation to the admissibility of the evidence of Detective Sergeant Smith concerning his recognition of the voice of the appellant after listening to approximately 15 seconds of the electronic record of one of the interviews conducted with the appellant at Clondalkin Garda station on the 27th or 28th of November 2008. The evidence given in the voir dire concerned the circumstances in which the voice recognition or voice identification evidence was obtained. Once the controversial evidence was ruled admissible pursuant to the learned trial judge’s ruling the same evidence was then placed before the jury, and is included in the summary above. 29. The learned trial judge was asked to exclude the controversial evidence on the basis: (1) that the prejudicial effect of the evidence was said to outweigh its probative value and (2) that there was an inherent unfairness about the manner in which the evidence in question was gathered and presented. 30. As to the first of these points it had been submitted that the prejudicial effect of the evidence outweighed its probative value in circumstances where the witness, Detective Sergeant Smith, was not in a position to identify any particular features of the speech or voice pattern of the person who made the threatening call, and he was merely asserting that he was capable of identifying the voice of the appellant with reference to that call. Moreover, he was not in a position to identify the section of the tape of interview or, for that matter which tape of interview, had been played for the purposes of his identification. It was contended that in the circumstances little or no weight could attached to this evidence and that its probative value would have to be regarded as effectively non-existent, or at most very slight. As against that, it was potentially very damning and prejudicial of the defendant. It was suggested that if it was allowed to go to the jury there was a risk that unjustified or inappropriate weight might be attached to it, and that then, when weighed in the balance with other evidence, it could cause the metaphorical scales of justice to tip against the defendant. It was submitted that in the interests of fairness the learned trial judge ought to have ruled the voice recognition evidence inadmissible, 31. The basis of the second complaint was that no procedure analogous to that used for formal visual identification, i.e., a visual identification parade, was either considered or attempted. It was suggested that a voice identification parade, or some other voice identification procedure involving foils for the purposes of comparison ought to have been conducted. It was further suggested that in any such voice identification procedure recordings ought properly to be maintained of the suspect’s voice sample, and of the other voice sample or samples used in the exercise as foils for the purposes of comparison to facilitate identification. It was further suggested that voice samples used as foils should be chosen on the basis of having some degree of similarity with the suspect’s voice, e.g. in terms of accent. 32. In reply, on the first point, counsel for the prosecution argued that the evidence was prejudicial precisely because it was probative. There were potential infirmities in relation to all forms of identification evidence. He conceded that in the circumstances the learned trial judge would be required to give a Casey / Turnbull type warning, suitably modified and adapted to the context of voice identification evidence. However, he submitted, as the evidence was relevant, and potentially probative, it should go to the jury and it was a matter for the jury, having been appropriately warned by the judge, to attach such weight to it as they thought fit. 33. On the second point, counsel for the prosecution noted that no authority had been cited in support of the proposition that a voice identification parade of the type suggested by the defendant was required before voice identification evidence could be fairly admitted. The suggested procedure was unprecedented and had not been mandated by any court in so far as he was aware. He submitted that the learned trial judge was entirely capable of charging the jury in a manner which met the requirements for justice and fair play. He was conceding that the jury ought to be given a Casey / Turnbull type warning, suitably modified and adapted to the context of voice identification evidence. That, he submitted, would meet the problem. 34. The learned trial judge ruled as follows at the end of the voir dire: “JUDGE: All right. Thanks, Mr Bowman. Well, I think Mr Bowman's application really has two legs to the stool, so to speak. Firstly, that it should be, that is the evidence proposed by Sergeant Smith, should be regarded as inadmissible on the basis that its prejudicial effect outweighs its likely probative value. For the reasons I think that we have already alluded to, I disagree with that. It is probative of a significant issue in the case and of course is intended to be prejudicial and this is not a case where a snippet of evidence has very little probative value of any issue in the case if accepted but has massive overtones of prejudice. So, I don't feel, rightly or wrongly, I don't feel that it can be eliminated on that basis. The second and more intriguing leg of his argument is that it should be basically eliminated on the basis of an absence of fairness in terms of the manner in which the evidence was gathered and presented and that is that, in effect, some type of procedure analogous to a visual identification parade, an audio parade perhaps should have been conducted in the manner suggested. That's a very interesting proposition; and although it appears that this issue has been considered to a very limited extent many years ago here, the issue of audio recognition apparently has been considered by the English courts and they have apparently mandated a very strong warning, perhaps even stronger than visual identification. Nowhere is there authority for the proposition that the kind of procedure that Mr Bowman suggests should be carried out. That's not to say that it doesn't make sense and it's not to say that Mr Bowman can't ultimately put it forward to the jury in terms of assessing the weight to be given to evidence such as this and it might indeed be instructive to the jury and prevent them from jumping to easy conclusions based on very brief recognition evidence, brief in the sense that it's based apparently on the sergeant being happy after 15 or so seconds of listening to a video of or the audio portion of a video where Mr Crowe is speaking being satisfied by way of comparison to something that he heard over a mobile phone some 24 hours previously. It might place it very much in context and it might reduce the jury's confidence in the matter whereby they wouldn't be prepared to accept the matter beyond reasonable doubt. But so far as the law as currently constituted is concerned, well, there isn't any law on that proposition. It's a very interesting proposition, but I can't feel that it renders the matter inadmissible. It certainly raises considerations which go to weight and certainly it would appear that the jury have to be warned very strongly before they could consider accepting such evidence beyond reasonable doubt. But in the circumstances, I think that the evidence is admissible, that weight is a matter for the jury.” 35. Subsequently, in charging the jury, the learned trial judge gave them the following warning in relation to the voice identification evidence that they had heard: “There is one specific legal issue I do want to dwell on. It's a case which depends partially but not wholly but very substantially on voice identification, members of the jury, and I want to give you the warning in relation to that to take with you into your deliberations and to apply in the course of your deliberations and it's identification rather than recognition. I mean, Sergeant Smith did say that he had spoken on a couple of occasions previously to Mr Crowe but in no sense was he purporting to place his identification of the voice on recognising it as a person that he had previously spoken to and knowing who he was speaking to. What he is doing is saying I heard 15 seconds of a voice on the audio portion of a videotape recording and he is saying to you, "I believe 100 % that that is the voice that I heard some 24 hours or more previously in the context of a mobile phone conversation or exchange", or communication perhaps more properly so called. So he's not recognising it, he's identifying it and he's identifying what he says he heard on the tape and as being what he heard in the mobile phone communication a day or so earlier. Now, identification whether it's by ear or by eye, members of the jury, is a human process and you'll all know yourselves the fallibility of human processes, including the fallibility of identification processes and I want to give you a warning in relation to that and it's a warning that arises in relation to responsible witnesses whose honesty is not in question and, indeed, witnesses whose opportunities for --observation is visual, but for listening had been adequate. People who are honest, and I don't think anybody is suggesting that the sergeant is dishonest, but whose honesty is not in question and who've had ample and adequate opportunities to hear what they say they heard, but people in those circumstances have made positive identifications which have subsequently been proved to be erroneous, wrong, mistaken, members of the jury. So, accordingly, I'm giving you a warning that you should be especially cautious before you accept such evidence of aural, A U R A L, identification as being correct, but if after careful examination of such evidence and this is important in the light of all of the circumstances and in the light of all of the other evidence in the case -- I mean, you examine it on its own but you also examine it in context, members of the jury, and with due regard to the other evidence in the case, if you are satisfied beyond reasonable doubt of the correctness of the identification, then you are at liberty to act upon it. So it's not a prohibition, it is a warning to be especially careful before you act upon such evidence, to consider all of the circumstances and the context of the recognition evidence and to consider it with due regard to all the other evidence in the case. It's to be cautious but it's not a prohibition and to be especially cautious and to be perhaps more especially cautious than you even would be in a visual identification case. That's what the English courts have said and the Irish courts haven't said otherwise, that a particularly strong warning, even over and above what would be required in visual identification, should be given in cases such as this. So I'm giving you that strong warning, members of the jury, and I want to contextualise the warning for you in relation to the particular recognition or identification which was carried out by Sergeant Smith. And this is non-exhaustive; you can use your own facilitates and your own judgment in deciding which aspects of it might be relevant. I'm simply contextualising the bits that I've seen and made a note of. You should look fist of all at the original communication and the circumstances of that, the time of the morning, where the officer was at the time, that it was communication over a mobile phone line, it was a communication of 30 seconds duration and you heard how much was on one side and how much was on the other in terms of the conversation. You should consider the kind of communication that it was in the context of the subsequent identification a day or so later, members of the jury, you should consider that that was based upon a 15 second excerpt or some time not much longer than that. You should consider that the sergeant didn't purport to base it on any particular characteristics of the voice that he heard on the video or the audio part of the video replay. You should bear that in mind. He wasn't saying that it was a particular tone; it was a particular accent or a particular type of accent or a particular region of a particular part of the country. He didn't give any features. He just said I heard 15 seconds and I'm satisfied of the 15 seconds I heard on the audio replay of the tape was the same voice as I heard on the mobile phone line 24 hours or so earlier and you can look at the differences in relation to that and take into account whether they are significant or insignificant. I'm simply highlighting the warning I given you -- I gave you, am giving you and trying to contextualise it for you. You should also bear in mind that notwithstanding the fact that he said that he had spoken to Mr Crowe a couple of times previously, he didn't say that he recognised him when he first heard the voice on the mobile phone and he wasn't basing his subsequent identification on that. He is expressly factoring that out of the equation and he didn't purport to recognise Mr Crowe either at the time or, indeed, subsequently. He said he was marrying up what he heard on the videotape with what he had heard on the mobile phone line, members of the jury. It's for you to look and to be very careful at all aspects of that and to ask yourselves are you satisfied beyond reasonable doubt to act upon the evidence bearing in mind the need for special caution in the case. Another thing that perhaps should feed into whether or not you are willing to act upon it is that there wasn't convened the equivalent of a visual identification parade, which I suppose the equivalent would be playing six recordings or whatever number of different people who perhaps come from the same broad area of the city as the person being purported to be identified. That there was no parade in that sense and that is the fact. It doesn't mean that you can't act upon the evidence if you are satisfied beyond reasonable doubt, but I'm simply saying that a parade carries with it input or potential input from the accused person because either they, or perhaps with their solicitor present, they can say, "Well, no, I'm objecting to that voice because it's not sufficiently similar", or, you know, that kind of thing. But also that if somebody recognises or identifies a voice from six or seven, you might be more comfortable in saying, "Well, you know, he had a lot of comparisons and he came up with the right answer". It might perhaps help you to be more certain, more satisfied beyond reasonable doubt than simply basing an identification on one snippet of 15 or so seconds rather than on a variety. So that's something you should bear in mind. You should also bear in mind the circumstances of the identification and that is, I mean, I know the sergeant is an experienced police officer and all of that, but, I mean, he was being brought in as a witness, not as a police officer, but as a witness who happened to be a police officer and to listen to something, but it's done so in the context of a suspect who'd been interviewed and that being replayed back. It's a bit like a dock identification where somebody in court says, "I'm identifying the person in the dock as the person who assaulted me on that particular day". Well, dock identifications are used very sparingly, members of the jury, for the obvious reason that the person who is in the dock is sitting in a special place in court and although they're accused and presumed to be innocent, there's a tendency perhaps to say, "Well, they're sitting in the docks, they must be the one who did it", and an equivalent tendency might be at work in even a sergeant carrying out -- an experienced sergeant, carrying out aural recognition in the circumstances in which he did, members of the jury. So I'm putting that into the equation to say it was a very informal type of procedure rather than a formal procedure and you should look at that aspect of it before deciding that you can be satisfied beyond reasonable doubt. But if at the end of the day, bearing in mind all of those matters, you are satisfied that both within itself and having due regard to all of the other evidence that you can act upon it beyond reasonable doubt, well, then, you're at liberty so to do. I'm simply obliged to give a strong warning and to contextualise it and to place it in the context of more formal procedures that might have been adopted but weren't, members of the jury. But if notwithstanding all of that you're satisfied of the correctness beyond reasonable doubt of the identification you're perfectly at liberty to act upon it, members of the jury.” 36. Neither side raised any requisition in respect of this charge. The arguments before this Court 38. As had been the case before the court of trial, the fair procedures argument was constructed with reference to the somewhat analogous situation of visual identification evidence. Counsel for the appellant sought to further develop his basic submission in a number of respects. He submitted that it was well recognised that for many reasons identification evidence can be unreliable, and that for many years the courts have recognised this fact and have sought to take countermeasures to guard against the possibility of wrongful convictions based on unreliable identification evidence. Sometimes that is achieved by insistence upon the adoption of safeguards in the evidence gathering process, or by the giving of appropriate warnings to the jury, or by a combination of both. 39. Counsel submitted that while the jurisprudence concerning how identification evidence is properly to be approached is most highly developed in the case of visual identification evidence and visual recognition evidence, there was no reason in principle, and indeed fairness required, that an analogous approach be adopted to any other forms of non-scientific sensory based identification or recognition where there is good reason to be concerned about the risk of wrongful conviction due to error. 40. The Court’s attention was drawn to People (Attorney General) v Casey (No 2) [1963] I.R. 33 and to R v Turnbull [1977] Q.B. 224 as being illustrative of the general approach as it applies in the visual identification context. To those, the Court itself would add the more recent cases of People (Director of Public Prosecutions) v O’Reilly [1990] 2. I.R. 415 and People (Director of Public Prosecutions) v Beroket McKonnen [2001] IECCA 74. 41. Although counsel was unable to refer the Court to any Irish authority dealing with voice identification evidence specifically, the Court was referred, without objection, to two academic papers dealing with the potential infirmities that may attach to such evidence, and also to the leading UK authority on voice recognition evidence i.e., the decision of the Court of Appeal (Criminal Division) in England in R v. Flynn and St John [2008] EWCA Crim 970; (2008) 2 Cr. App. R. 266. 42. The first paper to which the Court was referred was entitled “Sound Familiar? Voice identification evidence” by David Ormerod in 2001 Crim L.R. 595. It provides a detailed analysis of the dangers of using voice identification and voice recognition evidence and proposes safeguards for its use in the pre-trial and trial procedure. The paper examines inter alia voice parade procedures, and looks at single person parades, multi-person parades, parade safeguards, the use of foils, the question of what should the voices on parade be saying, the obtaining of voice samples, and issues surrounding the conduct of a voice parade such as size, whether voices should be taped or live, and replays. 43. The following passages from Mr Ormerod’s 2001 paper are of particular relevance having regard to the manner in which the voice identification evidence in the present case was gathered: “Given that voices must be tested sequentially, an obvious option would be to test the witness by requiring him to confirm the voice from a single person parade. With visual identification, the main dangers of single person confrontations are bias and suggestibility, and these pertain to voice identification because only one voice is heard. However, in practical terms, there is a clear advantage in using a single voice because it avoids the difficult task of compiling a parade of similar voices. Some research suggests that it might also lead to witnesses being more cautious in their selection. Although one-person parades have been used in the USA (e.g. in the much criticised case of Stovall v. Denno (1967) 388 U.S. 293) this is not, it is submitted, a desirable approach. In general, research demonstrates that the "use of one-person voice line-ups are likely to produce significantly more false identifications than the use of many-person line-ups (ref: Yarmey, Yarmey and Yarmey “Face and voice identifications in showups and line-ups” (1994) Applied Cognitive Psychology 453): in that experiment only 28 per cent were accurate. Although more cost efficient, the desire for accuracy and reliability must outrank mere practical gains. Moreover, in some jurisdictions one-person parades have been accepted in especially hazardous circumstances which deserve highlighting so that the same mistakes are not repeated. One particular technique is for the police to invite the witness to "overhear" a conversation occurring in an adjoining room. The witness will (usually) be alert to the fact that only one of the voices is that of a suspect, and from the content of the conversation will be readily able to distinguish that person from the investigators present e.g. U.S. v Pheaster (1976) 544 F2d 353. In addition, the investigators are unlikely to resemble the witness's description of the offender's voice; the suspect is.” 44. The second paper to which the Court was referred was “Sounding out expert voice identification”, again by David Ormerod, in 2002 Crim L.R. 771. This paper reviews expert evidence of voice identification, distinguishes between acoustic analysis and auditory analysis of voice, and contrasts the preference expressed for acoustic analysis by the Court of Appeal in Northern Ireland in R v O’Doherty [2002] N.I.L.R. 263 with the qualified approval given to auditory analysis by the English Court of Appeal in R v Robb (1991) Cr App. R. 161. Mr Ormerod’s 2002 paper is expressly referred to by Lord Justice Gage in his judgment in R v. Flynn and St John (at para 17). 45. Counsel for the appellant relies upon the case of R v. Flynn and St John as illustrating the need for safeguards and procedural fairness. In that case the court was concerned with both expert and non expert (referred to as lay listener) voice recognition evidence. It also had before it expert testimony concerning the potential infirmities attaching to non-expert voice recognition evidence. Counsel for the appellant in the present case has drawn this Court’s particular attention to paragraph 17 of Lord Justice Gage’s judgment where he states:
(1) Identification of a suspect by voice recognition is more difficult than visual identification. (2) Identification by voice recognition is likely to be more reliable when carried out by experts using acoustic and spectrographic techniques as well as sophisticated auditory techniques, than lay listener identification. (3) The ability of a lay listener correctly to identify voices is subject to a number of variables. There is at present little research about the effect of variability but the following factors are relevant: (i) the quality of the recording of the disputed voice or voices; (ii) the gap in time between the listener hearing the known voice and his attempt to recognise the disputed voice; (iii) the ability of the individual lay listener to identify voices in general. Research shows that the ability of an individual to identify voices varies from person to person. (iv) the nature and duration of the speech which is sought to be identified is important. Obviously, some voices are more distinctive than others and the longer the sample of speech the better the prospect of identification. (v) the greater the familiarity of the listener with the known voice the better his or her chance of accurately identifying a disputed voice. However, research shows that a confident recognition by a lay listener of a familiar voice may nevertheless be wrong. One study used telephone speech and involved fourteen people representing three generations of the same family being presented with speech recorded over both mobile and land line telephones. The results showed that some listeners produced mis-identifications, failing to identify family members or asserting some recordings did not represent any member of the family. The study used clear recordings of people speaking directly into the telephone. (4) Dr Holmes states that the crucial difference between a lay listener and expert speech analysis is that the expert is able to draw up an overall profile of the individual's speech patterns, in which the significance of each parameter is assessed individually, backed up with instrumental analysis and reference research. In contrast, the lay listener's response is fundamentally opaque. The lay listener cannot know and has no way of explaining, which aspects of the speaker's speech patterns he is responding to. He also has no way of assessing the significance of individual observed features relative to the overall speech profile. We add, the latter is a difference between visual identification and voice recognition; and the opaque nature of the lay listener's voice recognitions will make it more difficult to challenge the accuracy of their evidence.”
48. It is quite clear that the covert recording of voices in the Sprinter van was poor. It was so poor that Dr Holmes, even using her sophisticated equipment was unable to distinguish between different voices. Comparing her transcript with that of DC Gittings a number of the words which DC Gittings purported to recognise as Flynn's, are not distinguished by Dr Holmes as words. Professor French described the recording as poor and, as already stated, he was not able to rule out the possibility that the voice attributable to St John was either his or not his. 49. So far as the evidence of DC Gittings and DC Fleck is concerned, their familiarity with Flynn's voice was gained from comparatively short periods of time in his company on two days. It was eighteen days before DC Gittings listened to the covert recording and compiled the transcript. In DC Fleck's case, accepting his evidence which is in complete conflict with the Gittings evidence as to when the transcript was made, it was at least nine days after he had last heard Flynn speak before he listened to the covert recording. 50. So far as St John is concerned, the police officers listened to the covert recording very soon after they had been with him for a period of two hours twenty minutes. To that extent they were in a better position than DC Gittings and DC Fleck to identify his voice on the covert recording. However, DC Nicoll and DC Seymour had a limited time to familiarise themselves with St John's voice. Furthermore, the part of the covert recording alleged to be attributable to him lasted only 59 seconds of a poor tape and was of a voice itself said to be communicated to the van via the further distorting medium of a walkie-talkie. 51. There are other considerations which apply to both appellants. We have already noted the two most important factors, namely the limited opportunity for the officers to acquire familiarity with the appellants' voices and secondly, the poor quality of the covert recording. To these we add the following. First, the police officers' purported recognition of the appellants' voices is in marked contrast to the evidence of the two experts who are unable to recognise their voices; in Dr Holmes' case she was unable to identify individual voices. Secondly, nothing is known of the ability of any of the police officers to recognise voices. There is no evidence that any of them had any training in auditory analysis. Thirdly, the identification of the voices was carried out by listening to the covert tape on a standard laptop computer, as opposed to the sophisticated equipment used by the experts. Fourthly, in our judgment, it is significant that DC Gittings' transcript contains words attributed to Flynn which Dr Holmes could not distinguish as words, let alone recognise as attributable to Flynn. Fifthly, the expert evidence shows that lay listeners with considerable familiarity of a voice and listening to a clear recording, can still make mistakes. 52. Taking all these factors into account we conclude that there are powerful factors militating against the admission of the evidence of the police officers in the case of each of the appellants. Such evidence was self-evidently very prejudicial. As the experts point out, there are difficulties in challenging the evidence of lay listeners for the reasons noted earlier in this judgment. In our judgment, the general uncertainties about the evidence of voice recognition by lay listeners are enhanced by the specific facts in the case of each of these appellants. Quite simply, in our opinion, the prejudicial effect of this evidence far outweighed its probative value. Accordingly we conclude that the judge was wrong to rule it admissible. 53. There are other reasons why, in our judgment, the judge ought also to have excluded the evidence under s.78 of PACE. First, in our opinion, when the process of obtaining such evidence is embarked on by police officers it is vital that the process is properly recorded by those officers. The amount of time spent in contact with the defendant will be very relevant to the issue of familiarity. Secondly, the date and time spent by the police officer compiling a transcript of a covert recording must be recorded. If the police officer annotates the transcript with his views as to which person is speaking, that must be noted. Thirdly, before attempting the voice recognition exercise the police officer should not be supplied with a copy of a transcript bearing another officer's annotations of whom he believes is speaking. Any annotated transcript clearly compromises the ability of a subsequent listener to reach an independent opinion. Fourthly, for obvious reasons, it is highly desirable that such a voice recognition exercise should be carried out by someone other than an officer investigating the offence. It is all too easy for an investigating officer wittingly or unwittingly to be affected by knowledge already obtained in the course of the investigation. 54. This case provides an example of why these minimal safeguards should have been observed.” 48. Counsel for the appellant further submitted that were this court to hold that the voice identification evidence in the present case had been properly admitted before the jury it would, in effect, be setting the bar so low in terms of procedural fairness, that no voice identification or recognition evidence would ever be excluded. The appellant’s expectation was that the courts would scrutinise what occurred to see if it met minimum standards of fairness. It was submitted that in accordance with People (Director of Public Prosecutions) v Shaw [1982] I.R. 1 evidence should be excluded if, by reason of the manner in which it was obtained, it falls below minimum standards of fairness. The appellant’s case is that the voice identification evidence in this case was obtained in circumstances that fall below minimum standards of fairness, and that the learned trial judge ought to have excluded it. 49. Counsel for the appellant points out that in the context of visual identifications the Courts have long made it clear that the holding of an identity parade is by far the preferred option. Where a formal identity parade cannot be conducted, for want of co-operation by the suspect, or for other good reason, a more informal procedure may be permitted, but only providing that minimum standards of fairness can be achieved. While counsel for the respondent has sought to argue that all identification evidence, including dock identifications, is admissible in principle, subject only to weight, counsel for the appellant disputes this proposition and says that such evidence must be excluded if it fails to meet minimum standards of fairness. 50. Counsel for the respondent’s contention that dock identification evidence is still regarded as admissible and acceptable in certain circumstances was based upon his interpretation of the judgment of the Court of Criminal Appeal in People (Director of Public Prosecutions) v Cooney [1997] 3 I.R. 205. In the Cooney case a dock identification was indeed ruled admissible, but it was in circumstances where an identity parade had in fact been held but the evidence obtained as a result of it was excluded because the parade had been held while the accused was in unlawful custody. Counsel for the appellant contends that the Cooney case was decided on its own peculiar circumstances, and he significantly disputes that it is authority for anything other that the proposition that dock identifications are undesirable and unsatisfactory in general, as was stated by Keane J in his judgment on behalf of the court. 51. A further facet to the appellant’s case, in so far as it depends on allegedly unfair procedures, is that the electronic recording used in the identification was so used in breach of regulation 17 of the Criminal Justice Act 1984 (Electronic Recording of Interviews) Regulations, 1997, SI 74/1997, which provides:
(a) producing a photograph, or (b) otherwise for the identification. of the person interviewed except with the written consent of that person.” 53. The breach of regulations point was not made before the court of trial. However, counsel for the respondent has indicated to this court that he is taking no point on this. Notwithstanding that, he disputes the applicability of regulation 17 in the circumstances of this case. The respondent accepts that regulation 17 prohibits the use of electronically recorded interviews for the purposes of identification save where there is a signed consent from an accused. However, he submitted that this refers, in the ordinary way, to visual identification and is by no means intended to exclude their subsequent use in voice identification. 54. It was further submitted that in the UK and Ireland, visual identification procedures are subject to an elaborate regulatory framework although there is nothing similar for voice identification. It was submitted that there was no requirement to obtain written consent from the appellant and, consequently, there was no breach of the regulations. 55. Counsel for the respondent argues further and in the alternative, that, even if a breach of the regulation had occurred, this Court still has a discretion to excuse the breach under s. 7 (3) of the Criminal Justice Act 1984, which states:
The Courts decision 58. It is fairly criticised, firstly, on the basis that Detective Sergeant Smith knew that the person being interviewed (i.e., the appellant) was the only suspect in the case, and may on account of that have been subliminally biased towards a positive identification; secondly, on the basis that his identification was not based on any specific identifying characteristics (such as accent; timbre of voice; an attribute or trait such as a lisp, a stutter or a stammer; usage of colloquialisms or idiosyncrasy in manner of expression; or an individual mannerism such as a reflex clearing of the throat before commencing speaking) thereby rendering it difficult to test its reliability in cross-examination; thirdly, because the identification procedure lacked safeguards such as the use of foils for comparison purposes, and, fourthly, because no record was kept concerning which of the appellant’s recorded interviews, and what exact portion of the recording in question, comprised the 15 second segment of that material which, when it was played to Detective Sergeant Smith, allowed him to identify the appellant’s voice as being the same as the voice that he had heard on the telephone. 59. That having been said, can it be said that the evidence in question lacked any cogency at all? Detective Sergeant Smith made his identification just the day after he had received the phone call, when his memory was very fresh. His evidence was that the content of the call was threatening and menacing and that, notwithstanding that speaker’s voice was nondescript, it had resonated immediately with him. His identification of the voice was almost instantaneous, i.e., after listening to just 15 seconds or so of the recording, and was expressed with certitude. When pressed in cross-examination as whether he could be mistaken, Detective Sergeant Smith insisted he was 100% certain as to the correctness of his identification and that the chances of error on his part were zero. 60. The Court’s assessment is that while the controversial evidence is not utterly devoid of cogency, its cogency is minimal in the circumstances in which it was obtained in this case. 61. For a conviction based solely on identification evidence to be considered safe the evidence in question, notwithstanding possible infirmities, must above all have significant cogency. Where such cogency exists an appropriate warning of the Casey / Turnbull type drawing the jury’s attention to the inherent dangers in relying on such evidence, and to the specific infirmities in the identification evidence adduced before them, will be generally be a sufficient safeguard against the possibility of wrongful conviction. In the Court’s view if the only, or indeed the main, evidence against the appellant had been the voice identification evidence obtained in this case, the Court would have been unable to uphold the conviction based upon it for lack of sufficient cogency. 62. However, where a conviction is based in large measure upon circumstantial evidence, of which identification evidence represents but one strand, the required degree of cogency may be less than that required when such evidence is the sole or main evidence supporting the prosecution. The classical statement concerning the operation of circumstantial evidence is that of Atkins L.J in Thomas v Jones [1921] 1 K.B. 22, where he stated:
64. While the jury also heard the testimony of John Byrne and Kevanne McNamara, respectively, both of whom gave evidence on behalf of the appellant, the jury had the opportunity of observing those witnesses being tested as to their credibility and reliability in a rigorous cross-examination. It was expressly put to those witnesses that they had concocted a story and had given false testimony in an effort to assist the appellant. While both denied concoction and perjury, it is manifest from the transcript that neither witness withstood cross-examination well. The jury, in considering the totality of the evidence in the case, was entitled to reject the account given in court by John Byrne and Kevanne McNamara, as they manifestly did. 65. The issue for the Court is whether, in this particular case, based as it was on circumstantial evidence, the voice identification evidence was properly admitted before the jury as one of many circumstances in the case? The voice identification evidence in this case had some cogency, but its cogency was slight, particularly on account of the fact that Detective Sergeant Smith knew that the appellant was the only suspect in the case, thereby creating the risk of subliminal bias. It was also obtained in circumstances where there were no safeguards in place to counteract possible biases and errors, and accordingly it was weak and liable to significant criticism. These considerations beg the further question: did the voice identification evidence have enough cogency to justify it being admitted before the jury, always providing that the jury would be appropriately warned as to the risks associated with relying on such evidence in general, and providing that their attention would be drawn to the specific weaknesses and infirmities attaching to the particular voice identification evidence that they were being asked to consider in this case? 66. This Court has arrived at the unavoidable conclusion that the controversial evidence lacked the necessary degree of cogency to allow it to be admitted before the jury, even as a piece of circumstantial evidence. The Court has considered the Casey / Turnbull warning, adapted to the context of voice identification evidence, given to the jury in this particular case, and considers it to have been impeccable. However, in circumstances where the evidence to which the warning related failed to reach a minimum threshold of cogency, sufficient to allow the Court to conclude that a Casey / Turnbull type warning could neutralise any unfairness in the way in which the identification was carried out, and safeguard against the possibility of wrongful conviction, no warning of that type no matter how comprehensive or well expressed could suffice. 67. The Court is reinforced in its view that minimum standards of fairness were not achieved in circumstances where it is satisfied that there was indeed a breach of regulation 17 of the Criminal Justice Act 1984 (Electronic Recording of Interviews) Regulations, 1997. The Court considers that, having regard to the overall unfairness of the voice identification procedure engaged in, this is not an appropriate case in which to exercise its discretion under s. 7(3) of the Criminal Justice Act, 1984 to forgive the breach. 68. The Court is not impressed with the complaint that use of the recording of the appellant’s voice without his consent breached his privilege against self incrimination. The Court notes with approval the decision of Appeal Court of the High Court of Justiciary in Scotland in McFadden and Spark v Her Majesty’s Advocate [2009] HCJAC 78 where it was held that requiring a person to speak did not amount to a requirement that he incriminate himself. The Court held:
‘The court has consistently held...that the right not to incriminate oneself is primarily concerned with respecting the will of an accused person to remain silent. As commonly understood in the legal systems of the Contracting Parties to the Convention and elsewhere, it does not extend to the use in criminal proceedings of material which may be obtained from the accused through the use of compulsory powers but which has an existence independent of the will of the suspect such as, inter alia , documents acquired pursuant to a warrant, breath, blood, urine, hair or voice samples and bodily tissues for the purpose of DNA testing. . . [emphasis added]’” 70. The court considers that counsel for the appellant was correct in acknowledging that there may be no perfect or optimum system, and accordingly it is not for this court to be prescriptive concerning voice identification procedures in future cases. This is the first such case to come before this court and the absence of safeguards in this case, though a matter to be deprecated, may be partly explained by the novelty of the situation presenting itself to the investigating Gardaí in this case. Be that as it may, the Court agrees with counsel for the appellant that the total absence of safeguards meant that minimum standards of fairness were not met in the circumstances of this particular case, and accordingly the conviction cannot be upheld. 71. While it has not been prepared to do so in this particular case, the Court does not foreclose on the possibility that in a future case it might be persuaded to uphold the admission of voice identification evidence in the absence of appropriate procedural safeguards, depending on the circumstances of the case, and in particular the degree of procedural unfairness involved, and whether the attendant risk of wrongful conviction was capable of being adequately offset by other measures such as a Casey /Turnbull type warning. 72. Undoubtedly, the adoption in a particular case of a voice identification procedure which attempts to address potential biases and infirmities by means of safeguards, is likely to improve the cogency of such evidence. Therefore such measures are strongly to be encouraged on that account alone. Perhaps even more importantly they are also to be strongly encouraged in the interests of procedural fairness. 73. While there is no published guidance in this jurisdiction, there is abundant guidance in other jurisdictions to which recourse can be had. While not wishing to endorse any specific approach, the Court notes that in the neighbouring jurisdiction of England and Wales the Home Office has published a circular, No 057/2003 entitled “Advice on the use of voice identification parades”; while in Scotland the Lord Advocate has issued “Guidelines to Chief Constables on the conduct of visual identification procedures” which contains within it (at Appendix G) specific guidance for the conduct of identity parades for voice identification only. 74. In addition, the Court expects that experience gained in this jurisdiction with the more commonly encountered visual identification evidence, and the procedures and safeguards developed over many years in that context, should to a significant extent inform and influence the approach to ensuring fairness in any voice identification exercise that is to be attempted. 75. The appeal is allowed. |