Voice is caption: television interpreting and quality issues

Language is not the only way to execute an interaction: in oral communication much information is conveyed through extra-linguistic elements (Straniero Sergio 2007: 303), proxemics, gestures, voice and pauses (Cecot 2001). Stress and voice accentuation are important prosodic elements used by the interpreter as a caption, to convey information (Ahrens 2002: 42) and render the speaker’s “communicative intentions” (Kopczynski 1994: 89) to elicit a response from the audience. In the oral text voice intonation is also important in terms of cohesion, which is intended as “the result of the relationship between wordings, that is to say how the messages are expressed, and soundings, that is to say how they are uttered” (Straniero Sergio 2007: 300). In a “sequential environment” or in a “triadic exchange” the interpreter’s voice and intonation resources capture (Captionem) the listener’s attention and contribute to determine the progression of dialogic interaction (Straniero Sergio 2007: 301), facilitating the disambiguation process (Schlesinger 1994: 231). Therefore interpreters are evaluated not only for interpreting a speech correctly but convincingly well (Straniero Sergio 2007: 172). Interpreting quality is not an absolute value, but rather “contextually determined” (Kopczynski 1994: 87; Chesterman 1993 ). For the purpose of this study, quality can be defined as a pragmatic property where the interpreter “uses language in order to produce a particular effect in the mind of the listener” (Schlesinger 1994: 87) and on the “perceptions of the audience as users of the interpreters’ product” (Kurz 2001: 128). The study has been conducted on a small corpus of sub-corpus of 4 interviews for a total of 192 minutes in the language combination English-Italian, interpreted in simultaneous voice-over mode by a professional conference interpreter working for the Italian television. Both source and target texts have been transcribed (Straniero Sergio 2007) and analyzed (Heritage 1985) in order to detect the most significant prosodic features for quality assessment.