R v Nehme

Supreme Court of NSW|2024-03-15|Before: McNaughton J

JUDGMENT

On 18 March 2024, during the course of the presentation of the Crown case before the jury, I was asked to rule on the admissibility of a very short but contested audio passage ("the impugned passage"), in one of the many telephone interception recordings in this case. The impugned passage was part of a telephone call on 8 December 2019, beginning at 9:33am between Joseph Nehme ("the accused"), and another man, Mr Habib (Product 30080). The whole of this call was tendered on the voir dire on a USB.
I made two rulings: the first as to relevance, and the second as to whether the probative value of the evidence was outweighed by the danger of unfair prejudice. The rulings were made towards the end of the Crown case. Because it was important not to delay the progress of the jury trial, I provided a brief outline of the reasons for my decisions and indicated I would provide more detailed reasons later. These are those reasons.

Defence expert - Glenn Leembruggen

The defence tendered an expert report from Glenn Leembruggen, and also called him as a witness on the voir dire. He is the principal and managing director of an acoustic/audio consulting firm Acoustic Directions Pty Ltd. He has practiced as a consulting engineer since 1984, specialising in the fields of audio, electroacoustics and acoustics. He has undertaken post graduate studies in acoustics, taught electroacoustics and acoustics at the University of Sydney and has presented and published numerous technical papers in the fields of speech intelligibility, sounds systems and acoustics. He is a member of two international standards committees, one pertaining to the measurement of speech intelligibility.
He was provided with the audio recording of the impugned passage as well as the whole of the intercepted call contained in item 321. He was also provided with an enhanced version of the call, as well as a Crown Case Statement dated 28 April 2023, prepared by the Office of the Director of Public Prosecutions of NSW.
Mr Leembruggen reviewed the Crown Case Statement, and critically listened to the "original" and "enhanced" audio files using high quality, noise cancelling headphones. He used audio editing and processing software and applied a range of signal processing functions to both audio recordings in order to improve the voice clarity and analyse various properties of the recorded speech.
He noted the enhanced version had slightly degraded the speech clarity compared to the original recording because of a reduced high frequency content.
He analysed the impugned passage from four perspectives: First, he listened; second, he enhanced the recording and re-listened; third, he used software to compute the 'formants' in the impugned passage; and fourth, he compared the formants contained in the impugned passage to a recording of him saying "I stabbed a bloke" as well as to another recording produced by a synthesised male voice saying those words.
He said he could not hear the contended for words in either the original recording or the enhanced recording, even after using the Adobe Audition CC 2024 software by reducing the noise, adjusting the tone and slowing down the speech rate.
As to the formant analysis, he explained that a formant is a peak in the frequency spectrum of a specific utterance that results from an acoustic resonance within the human vocal tract. He used the software PRAAT for the analysis, which was developed by researchers at the University of Amsterdam.
Mr Leembruggen had already noted that the clarity of the accused's speech was degraded by factors including: his poor articulation of words, his higher than usual speaking rate, "Codec-induced artifacts" and "graininess" caused by the mobile phone transmission. He also acknowledged in cross-examination that the sound of the accused's motorbike would also diminish the transmission quality.
He said the formant analysis was consistent with the accused's words being very poorly formed, and with his inability to recognise any words in the impugned passage. He noted another part of the recording (not in contest) where the accused said, "why the fuck would I do that for?" showed a considerable movement in formants which coincided with the words being clearly discernible. He concluded the lack of formant change during the subject period would prevent reliable deciphering of specific words by listening. This adds weight, he said, to his conclusion that the contended for words were incorrect.
In comparing the impugned passage with recordings of the contended for words being spoken by him and the synthesized voice, Mr Leembruggen said they had no similarities.
In cross-examination, Mr Leembruggen agreed that the other phrase "why the fuck would I do that for" was spoken much more slowly than the impugned passage. He also agreed that he (Mr Leembruggen) was an Anglo-Saxon speaker, and that the synthesized voice sounded American.
In a question asked by me, he agreed that different dialects changed the formant structure because they have different sounds. He stated, "it's all to do with how we use our mouth to shape the words".
In further cross-examination by the Crown he stated that the steady formant sequence indicated there were no actual words, "[t]hey've been either destroyed by the phone transmission path or the poor articulation" (emphasis added).
He confirmed that he had not been provided with any other intercepted calls involving the accused, and he did not have the opportunity to get acquainted with the accused's particular dialect of English, nor was he given the opportunity to listen to at least two weeks' of intercepted calls involving the accused. He agreed that people can become familiar with someone's voice and manner of speaking, and context can assist understanding.

Crown witness - Detective Sergeant Colefax

The Crown tendered a statement prepared by the officer in charge, Detective Sergeant Colefax, and called him on the voir dire.
Detective Sergeant Colefax stated that he had prepared a transcription of telecommunications intercept material including, but not limited to, the transcripts tendered in the trial. The interception warrant was issued on 20 November 2019, and since that date, he reviewed most of the 6,587 voice call products, and created transcriptions or summaries of at least 609 products. At times he was assisted by others. In most cases, he would re-review the calls and summaries prepared by others as part of a quality-review process or to familiarise himself with relevant calls or the activities of the accused.
He said the initial review process involved him repeatedly starting, stopping and replaying the audio so that he could accurately summarise the call. At this stage of the review process, he listened to each call twice and then re-listened to difficult portions several additional times.
He also flagged calls or portions of calls that he may need to rely on again later and he would transcribe the relevant portions of the calls. This required further review of the relevant products.
Finally, he prepared court transcriptions which involved listening to each call again several times to consider its relevance to the prosecutions and then created a word-for-word transcription of each word spoken. This was a lengthy process given the need to accurately identify the words spoken and attribute them, sequentially, to each party to an intercepted call. This took some time. Detective Sergeant Colefax estimated one minute of an intercepted call would take between 10 to 30 minutes to transcribe.
Detective Sergeant Colefax prepared all the transcripts for court and became extremely familiar with the accused's voice and manner of speaking. The software he used assisted him and allowed him to monitor each call in stereo and isolate each side of the conversation as well as allowed him to alter the speed and volume of the call. He used high quality noise-cancelling headphones in an acoustic silent room. If he could not hear the words he would mark those portions "" which stands for indecipherable.
Detective Sergeant Colefax stated that the accused has a very distinctive voice in terms of its sound and projection. It was characterised by a distinctive hoarseness and is often croaky. There was a slight sort of Middle Eastern tone to it. Throughout the calls, the accused would vary his tone and projection. He would regularly lower his voice when discussing matters that were criminal in nature, such as what he referred to as "jobs". He said this quieter tone brought out the hoarseness in the accused's voice and was, at times, accompanied by increased pace as outlined later in relation to the impugned passage.
In relation to the impugned passage, Detective Sergeant Colefax stated he heard the accused briefly lower his voice, hurry his pace and say the portion emphasised in italics: "So do you know what I mean, but? I stabbed a bloke, do you know what I mean?".
Detective Sergeant Colefax stated that he recorded these words in the initial summary stage and through the subsequent transcriptions. He said despite the words beings spoken quickly, the words were clear to him. In oral evidence, Detective Sergeant Colefax stated that he had had 19 days' worth of listening by the time he heard the impugned passage. He noted that the accused was a very heavy user of his phone and made a lot of calls. A lot of them were lengthy in nature and were to a range of different people. He said he spoke in reasonable English, and, as noted, there was a slight sort of Middle Eastern tone to it. He also spoke in Arabic on occasions.
In cross-examination, it was suggested to Detective Sergeant Colefax that his hearing of the impugned passage had come about as a result of his involvement as the officer in charge, and from his understanding of the context of the material in the brief.

My listening to the impugned passage

Prior to coming to my decision, I listened to the audio recording a number of times. The first time was when it was played to me in Court. At that point I could not make out the words, nor had I been informed of what the Crown was contending were the words. After that, I was told that the words contended for were "I stabbed a bloke". Subsequently, I listened in Chambers to: 1. the audio recording of the contested passage provided to me on a USB and played by me multiple times using the sound system on my New South Wales Justice Department-issued computer; and 2. the slightly enhanced version provided to me, again using the same computer, to which I listened several times.
Finally, I note that I was also provided with a decision of Button J who was the presiding judge over the accused's trial prior to this trial, which, at that time, also involved four co-accused. That trial was aborted. In Button J's decision R v Nehme, Price, Rahim, Taufahema and Rizk (No 4) [2023] NSWSC 932, his Honour recorded that he had been asked to make certain pre-trial rulings, including being asked to exclude the impugned passage, the subject of this judgment. His Honour noted that his rulings were contingent (in the sense of being able to be revisited later in his trial).
Justice Button indicated that he had listened to the product himself in Chambers as the tribunal of fact on the voir dire. At one point he described the impugned passage as "pretty unclear and very ambiguous" and felt that he merely "might have heard" the alleged admission. He stated, "[h]aving said that, I pointed to the many 'potentialities' within the definition of relevant evidence in s 55 of the Evidence Act. In other words, to the extent that I felt that a person could interpret the sounds on the product in that way, my thought was that that crucial test could be passed". (Emphasis in original.)
Justice Button also expressed the view that, to guard against priming, the transcript of the impugned passage should not be provided to the jury. His Honour also noted that, because senior counsel foreshadowed the calling of expert evidence about the matter, he expressed himself tentatively. As noted above, in accordance with this ruling of Button J, the Crown does not press for the transcript of the impugned portion to be before the jury.

Consideration

The USB containing the recording of the impugned passage is a document, as defined in pt 1 of the Evidence Act Dictionary, it being "anything from which sounds […] can be reproduced with […] the aid of anything else".
Pursuant to s 48(1) of the Evidence Act the evidence of the contents of the USB may be tendered by tendering the USB itself. The evidence is the sound emanating from the playing of the USB on appropriate equipment.
Only if the sounds produced are relevant (Evidence Act, s 55), is the evidence admissible, unless they are to be excluded under other parts of the Evidence Act (s 56). The sounds produced are relevant, if accepted, they could rationally affect the assessment of the probability of the existence of a fact in issue in the proceeding (s 55).
In order for the sounds, if they are words, to be eligible as relevant in a court in New South Wales, the words need to be spoken in English (or so far as they were foreign words they would be those commonly used by Australian English speakers), and need to be capable of being understood.
I have considered the impugned passage myself, as set out above, together with the expert evidence called by the defence, and the evidence of Detective Sergeant Colefax.
I am of the view that Detective Sergeant Colefax can properly be regarded as an ad hoc expert in relation to understanding the words of the accused for the purpose of, at least, s 79 of the Evidence Act, given his extensive involvement with listening to the accused's voice over a period of many days which involved playing the calls multiple times. I leave for another day whether or not his evidence would be admissible on the voir dire also under s 78 of the Evidence Act.
As stated in Eastman v R (1997) 76 FCR 9 at 111, "The High court recognised that at times the sounds recorded on a tape may be indistinct and difficult to decipher. […] Mason CJ, Brennan and Deane JJ said [in Butera v Director of Public Prosecutions (Vic) (1987) 164 CLR 180; [1987] HCA 48 at 187]: 'Although evidence derived from a tape recording is not subject to some of the frailties of human testimony, it may exhibit deficiencies from which human testimony is usually free. A tape recording which is indistinct may not yield its full content to the listener on its first playing over. It may need to be played over repeatedly before the listener's ear becomes attuned to the words or other sounds recorded. This situation has led courts to receive transcripts not as evidence of the conversation or other sounds recorded but as a means of assisting in the perception and understanding of the evidence tendered by the playing over of the tape.'"
As stated in R v Menzies [1982] 1 NZLR 40 at 49, quoted with approval in Butera at 188: "Whether the Judge allows the jury to have copies of the transcript, as distinct from merely hearing it read, must be a matter for his discretion in the particular case, bearing in mind the requirements of justice and any risk of unfairness to the accused."
Even though I could not hear the impugned passage when I first listened to it, after I listened to it multiple times, and I got used to the speed and manner of the speaker, I have come to the view that, just like Button J, a person could interpret the sounds on the product in the way contended for, being "I stabbed a bloke". The slightly enhanced version made it slightly easier to hear as compared to the version I was initially given.
I also note that the words contended for, as emphasised in the oral evidence of Detective Sergeant Colefax, were heard by him during the first stage of him listening to the call, and consistently thereafter.
Further, in terms of context, it can be noted that the words contended for appear to align with the subject matter of the balance of the telephone call between the accused and Mr Habib in which they discuss what the police investigations may be able to determine as to links between the accused, Mr Habib's residence and the death of the deceased.
As noted, Mr Leembruggen was not able to discern the words contended for according to the various exercises he undertook. In my view, Mr Leembruggen was disadvantaged in that he did not have the opportunity to become familiar with the particular voice of the accused, which he himself acknowledged was poorly articulated, and in a particular dialect of English which he did not have the opportunity to get acquainted with. Further, the voices used by the expert for comparison did not, in my view, provide an appropriate comparator because they were not speaking the same dialect of English as the accused. I did not find the expert evidence persuasive.
I was referred to the judgments of N Adams J in R v Hawat (No 5) [2019] NSWSC 1727 and Johnson J in Decision restricted [2021] NSWSC 750, both of which I found most helpful. It can be noted that the calls excluded in those cases were only those calls where no sensible words could be discerned at all. Where enough words could be heard, both decisions allowed those conversations to be put before the jury. Ultimately, my approach accords with the approach of N Adams and Johnson JJ.
As the tribunal of fact on the voir dire, I am of the view that the words contended for are capable of being discerned by a jury, especially if they are listened to multiple times. My opinion, based on my own listening (which would have been enough to dispose of the relevance argument), is fortified by the evidence of Detective Sergeant Colefax. As indicated above, given the particular exercises undertaken by the defence expert, Mr Leembruggen, I did not find his evidence persuasive, and it did not undermine my own assessment.
Accordingly, in my view, the impugned passage of the transcript is relevant under s 55 of the Evidence Act in that, if the words contended for are accepted, they could rationally affect directly or indirectly the assessment of the probability of the existence of a fact in issue in the proceeding, namely, whether the accused stabbed the deceased or was closely involved with the stabbing of the deceased. Given the impugned passage is relevant, it is therefore admissible pursuant to s 56 of the Evidence Act unless excluded by another provision of the Evidence Act.

Second objection - s 137 of the Evidence Act

I then proceeded to consider, pursuant to s 137 of the Evidence Act, whether or not the impugned passage should be excluded on the basis that its probative value is substantially outweighed by the danger that the evidence might be unfairly prejudicial to a party.
Probative value is defined in the pt 1 of the Evidence Act Dictionary to mean the extent to which the evidence could rationally affect the assessment of the probability of the existence of a fact in issue.
The defence pointed to IMM v The Queen (2016) 257 CLR 300; [2016] HCA 14 at [50] ("IMM") in the High Court which included the following: "It must also be understood that the basis upon which a trial judge proceeds, that the jury will accept the evidence taken at its highest, does not distort a finding as to the real probative value of the evidence. The circumstances surrounding the evidence may indicate that its highest level is not very high at all. The example given by J D Heydon QC was of an identification made very briefly in foggy conditions and in bad light by a witness who did not know the person identified. As he points out, on one approach it is possible to say that taken at its highest it is as high as any other identification, and then look for particular weaknesses in the evidence (which would include reliability). On another approach, it is an identification, but a weak one because it is simply unconvincing. The former is the approach undertaken by the Victorian Court of Appeal; the latter by the New South Wales Court of Criminal Appeal. The point presently to be made is that it is the latter approach which the statute requires. This is the assessment undertaken by the trial judge of the probative value of the evidence." (Footnotes omitted.)
That example, in my view, is not analogous to the situation here because the evidence in this case is not being filtered through the visual or oral faculties of anyone else, as in the example in IMM where it was simply an account by a person of another occasion affected by various circumstances. Rather, in this case, there is objective evidence of a telephone call which can be placed before the jury for them to either be satisfied, or not, as to what they hear. If the jury listens to it and jurors find they cannot hear it, they cannot hear it. If they can, they can. It is thus a different type of evidence to that referred to in IMM.
True it is that the alleged admission, if it were found by the jury to have been said, is not unimportant. On its face, the probative value is high.
However, this is not to be viewed in a vacuum. This must be viewed in an important context, which includes the following. First, the Crown opened that the Crown did not know who the stabber was. In other words, the Crown is not relying on this alleged admission to prove that the accused was the stabber. This is perhaps understandable given that the accused both clearly does not always tell the truth during the intercepted calls and changes his account.
For example, there are instances where the accused has stated things confidently in the course of a telephone call which are patently untrue. For instance, he is recorded in a telephone intercept with Lisa Price stating that there was a child in the Lembryk/Bradley house, which is clearly not the case. At other times he gave an account that it was "they" who "thinged him", but another time he said "my mate got him", and another time he said "but he stabbed him". These are only examples I have pulled out in a short amount of time. In other words, the accused has made contradictory statements as to his direct involvement in the stabbing, and untrue statements about less central matters.
In my view, if it were the only admission and there were no examples of the accused giving false or contradictory accounts, the probative value would be arguably higher than it is. However, given the nature of the alleged admission in the context of the rest of the call, it is still of high probative value, as it clearly indicates that the accused regards himself as implicated in the stabbing.
I also note that the call is with Mr Habib (a man clearly associated with the address from which Viliami Taufahema was collected by Bilal Rahim prior to the killing) and the balance of the call clearly deals, in part, with the investigations that the police may have been undertaking and what links there may have been between the accused, Mr Habib's address and the incident. This context reinforces the probative value of the impugned passage.
As to the suggestion by the defence that the Crown should subpoena Mr Habib to give evidence as to what he understood was said on the call, it is a situation where, in my opinion, the jury can listen to the calls multiple times, and what Mr Habib may or may not be able to hear now is simply another version. Even though (as suggested by the defence) he may be more familiar with the accused's voice, any evidence he may give would, again, simply be another version of what could be heard. Because this is objective evidence and the jury can decide to accept or not accept it, and they will be firmly directed in relation to that, I am of the view that Mr Habib's evidence would not assist the jury in any significant way. In any event, whether or not the Crown calls Mr Habib, is clearly a matter for the Crown. In due course if a Mahmood direction is deemed appropriate, that can also be given to the jury. [1]
Here, if there is any prejudice, it would come from the issue of priming. Given the jury is not being provided with transcript, the only basis of the priming would come from the Crown's suggestion in its closing address. [2]
The suggested words would not have the imprimatur of the evidence of Detective Sergeant Colefax. They would not have the imprimatur of the jury having it before them in their folder (Exhibit F) along with all the other transcript in black and white. They would have the submission, and whilst they would be primed to a point, the suggested words would not be of an equal stature to all the other transcript which was before them by agreement.
To this can be added what the jurors have been told from the very beginning, more than once, that what they hear is the evidence, and what the transcript says should only be accepted by them if it accords with what they hear. Here, they will not even have the transcript. They will simply have the Crown submission [3] . This, in my view, minimises the issue of priming to a minimal level.
In summary, the probative value of the impugned call is high, but not as high as it would be in circumstances where it is the only admission. In any event, given the opportunity for priming is minimal and the evidence is otherwise able to be played multiple times, and will be subject to a reminder to jurors that it will be up to them to decide what they hear, the probative value is not outweighed by the danger of unfair prejudice to the accused. It is appropriate for the evidence to be placed before the jury. It is for the jurors to make up their own minds. I will give the members of the jury firm directions that they should only act on what they hear.
My decision is that the Crown is permitted to adduce the whole of the audio recording, including the impugned passage. The Crown will be permitted to address the jury as to the words contended for in its closing address. I will give the members of the jury appropriate directions in relation to the impugned passage that it is for them to make up their own minds as to what the impugned passage may or may not contain.

Endnotes

I note in passing that the High Court authority of Butera v Director of Public Prosecutions (Vic) (1987) 164 CLR 180; [1987] HCA 48, which characterised transcript (that is, English transcript) as an aide-mémoire rather than evidence, was handed down before the Evidence Act 1995 (NSW). This characterisation has endured beyond the introduction of the Evidence Act (see for instance R v Giovannone [2002] NSWCCA 323 at [62]). On further reflection, it may be more in keeping with the terms of the Evidence Act, to characterise the transcript as evidence under s 48(1)(c) of the Evidence Act, but to direct the jury pursuant to s 136 of the Evidence Act that the use of the transcript is limited to be of assistance to them only if it accords with what they hear, but to prefer what they hear over the transcript should there be any difference. Not only does this fit with the rubric of the Evidence Act (see R v Hall [2001] NSWSC 827 at [35]), it allows ss 135 and 137 to be engaged. Those sections are otherwise not engaged in relation to material which is not strictly "evidence" (see R v Cassar; R v Sleiman [1999] NSWSC 436 at [7]). Given, however, this point was not argued before me, I make no further comment about it.
Mahmood v Western Australia (2008) 232 CLR 397; [2008] HCA 1.
I note that since I delivered my oral remarks, the accused has given evidence, and the impugned words were put to him in cross-examination and denied by him.

Amendments 07 May 2024 - Representation amended. DISCLAIMER - Every effort has been made to comply with suppression orders or statutory provisions prohibiting publication that may apply to this judgment or decision. The onus remains on any person using material in the judgment or decision to ensure that the intended use of that material does not breach any such order or provision. Further enquiries may be directed to the Registry of the Court or Tribunal in which it was generated. Decision last updated: 07 May 2024

At a glance

Court

Decision date

Before

Catchwords

Source

Catchwords

Judgment (9 paragraphs)

Parties

Legislation Cited (1)

Cases Cited (13)

Cited By (1)