I am glad to see such a positive response to the editorial on interpreting practice. ^{Reference Cambridge, Singh and Johnson1} Psychiatry and speech and language therapy are two of the most challenging areas of practice for interpreters.

Australia has an honourable tradition in the field of language support for its diverse population, as I experienced in New South Wales a few years ago. Andrew Firestone's description of using a triangular seating arrangement but having changed to sitting the interpreter next to him is interesting. ^{Reference Firestone2} I have found that if I sit next to either the clinician or the patient, problems in the doctor-patient relationship can still occur. If closer to the patient, it is more likely that they will address questions directly to me, trying to draw me in ‘on their side’, such as ‘Are you married?’ or ‘Do you have children?’ If closer to the clinician, my impartiality can seem to the patient to be compromised.

In the UK almost all interpreters in the public sector are independent freelance workers. Being seen by the service user as directly employed by a state institution, whichever it is, can cause them to distrust our interpretation, especially if they have arrived from a totalitarian state. Seating the interpreter at the apex of an isosceles triangle, in which the clinician and patient are closest together and directly facing one another, allows eye contact to be maintained between them, and keeps the interpreter out of direct line of sight. Interpreters who are taking notes will be busy with their notebooks and not available for eye contact. They still need to be able to see the speakers’ faces, of course.

It would be interesting to know whether interpreters and clinicians maintain direct speech during clinic sessions, such as ‘How are you feeling?’ rather than ‘Ask her how she feels’. This is another way of keeping the interpreter out of a direct relationship with either party during the interview. It is very important that the interpreter introduces themselves and briefly explains how they work, at the beginning of the session. This, and everything else that is said, should be done in both languages. If the patient is reminded at the outset that ‘I will interpret everything I hear’ and ‘I will speak to you as the doctor does, with “I” and “you”; they are his words’, ownership of what is said remains with the primary interlocutors, not the interpreter.