It is a fundamental part of a criminal case to prove the accused is the person how has committed the offence otherwise identifying the offender. When the identification of the accused person’s is in question, it can be proved or disproved by adducing similarity or dissimilarity of the accused’s characteristics. And, one of the characteristics by which a person can be identified is his/her voice. If there is any relevant voice recording of an individual then the tape can be played before the jury to compare the voice of the accused and the voice of the offender. When, there is no recording of voice but someone heard the voice of the offender(s), it is solely depended on those witnesses to identify the accused(s) on the basis of their memory.
In relation to the cases of identifying the voice of the accused by the witness who heard the voice of the offender, the court has to mainly depend on the perception or the assertion of the witnesses regarding the voice of the offender. There is real possibility of mistake on expense of the surrounding circumstances when the witness perceived the voice. For example- the medium through which he perceived, the ability to remember the way the offender spoke, ability to compare the voices accurately, how long he heard the offender and the time passed between hearing the offender and the accused voice. The court held the non-expert or witness’s opinion is admissible depending upon ‘the degree of familiarity of the witness with the suspect’s voice’ as the strangers are likely to make mistakes. However, the danger of misidentification cannot be wiped out by the degree of familiarity and a stringent Turnbull direction to the jury will require than that for visual identification.[1]With visual identification the witnesses are the subject to an elaborate regulatory framework but nothing similar exists for the witnesses of voice identification.[2]
When there is any recording of the offender’s voice there can be two types of evidences:‘expert evidence’ and ‘lay listeners evidence’.[3]Experts can give evidence on the basis of auditory or acoustic analysis of the voice. In R v Hersey[4], a robbery case, a voice identification procedure had been carried out to recognise the accused. An expert gave evidence that the voices were pitch higher than the accused. However, the court allowed the identification evidence and ruled that the expert evidence was not admissible before the jury. The jury did not require the assistance of an expert where the matters in the question were within their own experience and competence, but a Turnbull warning should be given to them.
Since, R v Robert[5]the courts have acknowledged that the voice recognition is more difficult than visual identification of the accused and the premier should draw more warning than the later. Where a tape-recording is admitted in the evidence the expert evidence in phonetics is admissible to decide whether the voice matches that of the accused. Later, the court also held, without expert assistance the jury should not be allowed to compare the voice in recording with the voice in another of the accused.[6]However, it has been changed by the Court of Appeal in the case of R v Flynn[7].Now the jury is allowed to make a comparison between the voice in covert recording and the recording given by police as ‘lay listener’[8]identifying the accused.
Though, the court in this case stressed on the dangers of using police officers as ‘lay listeners’ to identify the voices, it does not provide enough guidance on dangers of mistaken identification, relevant factors to be considered by the judge regarding the quality of the voices.[9]There is no guideline to indicate the judge when to withdraw the case from the jury for the poor quality of voice and how to give them adequate warning for taking caution.
The Court of Appeal in R v Flynn provided some safeguards for pre-trial procedures.
· The evidence gathering procedure should be recorded, particularly the amount of time spent in contract with the accused when identifying the voice by an officer.
· The recording must be noted and annotated when compiling with the views of the offer as to who is speaking in the recording.
· The officer should not be supplied with a copy of another officer’s transcript where it contains his view of who is speaking.
· The identification should be carried out by someone other than an officer involved in investigation.
It is can be said that there are little safeguards toward the dangers of voice recognition. The Code D of Code of Practice for Police under PACE states, the code does not preclude police from using voice identification parade where they judge it is appropriate in the case. Later, this can be adduced to the court to prove the accused guilt. Though, the dangers of misidentification has been perceived by the courts for more than a decade and consequently outlined the safeguards necessary at pre-trial stage to minimise it, any of those has yet to come in the police code of practice. The only mechanism that exists now is the Turnbull warning which falls short of providing enough safeguard for voice recognition.
[1] R Emson, Palgrave Macmillan Law Masters Evidence (5th ed. Palgrave Macmillan, 2010), p.329.
[2] A Roberts, ‘Evidence: Identification – voice recognition’ [2008] 10 Crim LR 799-802.
[3] D Warburton & T Lewis, ‘Opinion evidence; admissibility of ad hoc expert voice recognition evidence: R. v Flynn’ [2009] 13 (1) E. & P. 50-57.
[4] R v Hersey [1998] Crim LR 281,CA.
[5] R v Roberts [2000] Crim LR 183, CA.
[6] R v Chenia [2004] 1 All ER 543.
[7][2008] Cr App R 20 at (56).
[8]Someone who possesses some special knowledge about the suspect, which enables him to recognise the suspect’s voice. Example- police or close relative.
[9] A Keane & others, The Modern Law of Evidence (8th ed. Oxford University Press, Oxford 2010), p.246.
No comments:
Post a Comment