John Wilding, Susan Cook and Josh Davis present some curious findings from the field of ‘earwitness’ testimony.
18 November 2000
In 1935 Bruno Hauptmann was convicted of the kidnapping and murder of the infant son of the aviator Charles Lindbergh in the United States. Lindbergh claimed that Hauptmann’s voice matched that of the man he had heard saying ‘Hey, doctor, over here, over here!’ when the ransom was paid three years earlier. This is the best-known example of voice identification in a legal case. There was virtually no relevant experimental evidence to aid the court, and the case prompted McGehee (1937) to undertake pioneering studies. However, 60 years later a report in The Guardian (5 September 1997) suggested that British courts and police forces were still very uninformed on the issues: …three Court of Appeal judges yesterday ordered a full hearing with leading counsel to explore the new police method of identifying suspects by ‘voice parades’. They adjourned yesterday’s hearing over a robbery conviction after the Crown’s counsel said police forces were anxious for some guidance over voice identification parades. Research on ‘earwitnessing’ is meagre and unsystematic compared with the data available on eyewitnessing in general and face identification in particular. Sophisticated models have been developed that incorporate both visual identification of the individual and the extraction of visual information relating to emotional expression and linguistic utterance (lip reading). The role of the voice in identifying the person and emotions is acknowledged by adding some boxes and arrows to the models and assuming that the component processes operate in parallel with the visual channels. We argue that more complex models of person identification are needed, which may in turn have a bearing on the interpretation of witness evidence.
Download PDF for article.