The author conducted a series of structured interviews of about 10 minutes with conference participants at OECD Headquarters meetings in Paris over a period of 15 months (January 2002 to April 2003) either during or just after the meeting. The author is aware of methodological pitfalls associated with an interpreter interviewing delegates about interpreting quality and endeavoured to reduce bias. She did not introduce herself as an interpreter, but said that the survey was being conducted by the conference service. Some of her findings are: - 9 respondents (24%) used interpreting because it is too tiring to listen to a foreign language all day. It seems that they use interpreting for comfort, even if they can get by without it. - Respondents felt able to judge the accuracy of the target language rendition by comparing notes with colleagues, on the basis of the coherence of the information served to them, of the interpreter’s tone, by the terminological accuracy of interpreting, etc. - Most of them seemed to take the interpreter’s accuracy for granted - Incorrect terminology was a frequently cited source of irritation and an indicator of lack of genuine understanding by the interpreter - In no case did an interviewee mention spontaneously correct pronunciation or accent of other formal elements of discourse - There was no indication that delegates had preferences as to directionality - Any prior disappointing experience with interpreting seemed to cause long-term damage in terms of loss of confidence. [Based on D. Gile]