CHAPTER THREE A TRAINING PROGRAM FOR EXPERT FORENSIC TRANSCRIBERS

Forensic linguistics is a discipline that concerns General Linguistics, in particular Applied Linguistics and Forensic Science. Today, it is an autonomous discipline with its own methodology and it can regard any text, written, recorded or produced orally, that is somehow involved in a legal proceeding or in a criminal context. “Literally any text or item of spoken language has the potential of being a forensic text” (Olsson 2008, 1). The forensic linguistics expert who analyses recorded spontaneous speech needs a number of different skills across several disciplines (Chaski 1998, 2001a, 2001b): Phonetics and phonology, necessary for the study of the articulatory sound system of human speech, the sound inventory which characterizes a language or a dialect and the rules that govern the adjustments and changes that occur in spontaneous speech1; Morphology, to study the smallest units of language with meaning, the organization and the study of the internal structure of words; Syntax, to be aware of the rules that govern the internal structure of sentences, how words can be combined to create linguistically acceptable sentences; Semantics, the study of meaning, the relationship between the meaning of (single) words and the meaning of (complete) sentences; Pragmatics, necessary to understand the use of language during a normal conversation and in all types of interactions and dialogue; Historical Linguistics, to study the evolution of a language and what the possible and predictable changes may be; Sociolinguistics, for the language used by a speech community, the study of the variables characterizing the language analyzed, focusing on differences of religion, social class, educational levels, gender, place of residence (city versus countryside, mountains versus coast), the interaction and interference between different languages and between language and dialect, and mixed-language production; Dialectology, to ensure a correct identification of dialects, local variables, regional dialects and their borders, the identification of isoglosses or isophones; Dialectometry, to study the distances between dialects and their internal regularity; Psycholinguistics, to understand how psychological and mental systems work in a human language; Neurolinguistics, to study the human nervous system and the neuroanatomical functioning of the brain; Computational Linguistics or Computer Science, to be able to use software and tools that help the expert during the analysis of the language and performance of individual speakers; Statistics, to analyse the results of a linguistic analysis, describe the phenomena and provide objective answers; Phonetics and Acoustic Signal analysis, to conduct an acoustic analysis of sounds, identify characteristic features, filter noisy signals, identify anonymous speakers, support the transcription of sound signals with low quality (Romito 2000); and, Law, to act in accordance with the procedural rules. The definition in The Routledge Encyclopaedia of Linguistics (2010) reports that “forensic linguistics helps courts to answer three questions about a text—what does it say, what does it mean and who wrote, typed or authored it?”—confirming the initial interest in the written text. The first study was made in 1930 (second edition Bryant 1962) and analysed the function of words in legal language. Forensic linguistics was first used in a courtroom to gather evidence in 1968, when Jan Svartvik (Svarvtik 1968) showed in The statements: a case for Forensic Linguistics how style and grammar could be measured, and thus become elements of evidence in a court case (Fitzgerald 2004). This court case and especially came to the attention of the world with the arrest of the Unabomber in 1996 in the woods of Montana2. The first analyses, therefore, took place on written texts. Nowadays, with the development of telecommunications, there has been an increase in analyses conducted on acoustic recordings rather than on written declarations. The analyses that can be carried out on recorded sounds are: the identification of a sound or a noise; the determination of the originality or authenticity of the tape, support or recording; speech signal filtering; noise reduction or emphasis of a voice and, in general, an increase in intelligibility; the identification of an altered voice; speaker characterization; a comparison and perceptual comparison of two voices; speaker identification by objective methods; the transcription of interceptions and analysis of meaning in relation to noisy recordings. Unfortunately, forensic linguistics or phonetics does not exist as a teaching course in Italian universities. It is impossible, therefore, to establish whether it concerns Criminalistics, Investigative Science or Applied Linguistics or even Human Sciences, as is the case in other countries. Furthermore, academic scientific research on issues related to Forensic Linguistics does not stimulate great interest. The enthusiasm registered in the US and the UK has gradually faded in Italy after an initial surge in the 1970s and 1980s. Experts do not cooperate with each other and they tend to be graduates, accountants, engineers, linguists, sound engineers, doctors and so on. A similar task has never been given to people coming from such diverse academic paths and backgrounds (Romito 2010; Romito and Galatà 2008; Romito et al. 2008). We believe that only an appropriate training course could develop competence and increase the confidence of judges towards this discipline.