Position Statement concerning use of impressionistic likelihood terms in forensic speaker comparison cases, with a foreword by Peter French and Philip Harrison

Over recent years a number of UK forensic speech scientists have become concerned about the framework in which conclusions are typically expressed in forensic speaker comparison cases. The issues for speech science came into sharp relief in the wake of an Appeal Court ruling (R -v- Doheny & Adams [1996]) which concluded that at an initial trial DNA findings were presented in a way that gave undue weight to the evidence. The particular approach to representing DNA evidence was ruled unlawful. On considering the impressionistic likelihood scales (French & Harrison 2006) used by most practitioners in speaker comparison cases it became clear that they gave the same false weighting to the speech findings as the outlawed framework did to the DNA findings. These views were first presented to the 14th meeting of the IAFPA in Marrakech, Morocco in August 2005 (French 2005). Following this, a group of speech scientists based in York held a series of meetings to discuss alternative approaches to expressing conclusions and bring the field into line with modern thinking in other areas of forensic science. A draft document was prepared with nput and assistance from Martin Duckworth (College of St Mark & St John, Plymouth). The contents were further developed and finalised as a position statement after a seminar involving the York group and a group from the University of Cambridge. The document appears below. In order to assess the level of support for the new position, the statement was circulated to all practising forensic speech scientists and interested academics within the UK. With one exception, all those contacted became co-signatories. The statement now reflects the all but unanimous position within the UK. A copy of the position statement has been lodged with the Crown Prosecution Service (CPS) Policy Division in the jurisdiction of England and Wales, with the office of the Procurator Fiscal in Scotland and with the office of the Director of Public Prosecutions in Northern Ireland. It has also been published on the internet (http://www.forensic-speech-science.info). To a very great extent the document speaks for itself. However, we take this opportunity to highlight the fact that the new approach brings about a fundamental change in the role of the analyst and the evidence. In the past forensic speech scientists were often thought of as identifying speakers. Within the new approach they do not make identifications. Rather, their role becomes that of providing an assessment of whether the voice in the questioned recording fits the description of the suspect. This assessment is not made in purely global or holistic terms in the way that a layman listening to the material might attempt to make it. Instead, it involves 'separating out' the samples into their constituent phonetic and acoustic 'strands' (e.g., voice quality, intonation, rhythm, tempo, articulation rate, consonant and vowel realisations) and analysing each one separately. It may also involve statistical analysis of the features found. In short, the rigour and detail of the analysis, together with the education, training and experience pre-requisite to carrying it out, put it well beyond the resources of a layman simply listening to the material. Additionally, by drawing upon research literature and general experience, the analyst may provide an assessment of the degree to which the features common to the questioned voice and that of the suspect are unusual or distinctive. In most cases, this is again beyond the knowledge and abilities of the layman. So, one might find, for example, that the voice in a criminal sample fits the description of the suspect and that the features they share are highly unusual ones. The ultimate decision as to whether they are the same person, however, does not lie with the expert. It is a decision to be made by the court (magistrate, judge, jury) and it is one they will make in the light of other evidence in the case. This new framework is, at a conceptual level, identical to that used nowadays in the presentation of DNA evidence. Neither the speech scientist nor DNA scientist dentifies the suspect. They provide the jury with the specialist scientific information necessary to make the identification (or otherwise) themselves. Finally, we would note that in formulating the arguments in the body of the position statement we have been at pains to avoid esoteric terminology wherever possible, thereby maintaining wide accessibility to non-specialists, including lawyers. The editors of the journal invite both positive and negative responses to this Position Statement for publication in subsequent issues.