Interpreters, Psychosis, and Voices

From the Psychology Department Newsletter June 2004.

Using an Interpreter

It is almost entirely true to say that in psychology if you can’t talk to your client you can’t do your job. It is certainly entirely false to say that with an interpreter (of any language) you can get on with your job ‘as normal’. The immediate impact of a third party is probably obvious – dynamically, temporally (with spoken language interpreting, being sequential), and transferentially (particularly with BSL interpreters where the client’s emotional relationship is often more powerfully with the interpreter than the clinician). Alongside these more immediately apparent difficulties lies a more pervasive problem of accurate communication, understanding, and - for want of a better term – empathy. While the relation between language and thought is a complex one, it is nevertheless fairly clear that expressed language in any modality is an encoded version of the concepts intended by the person speaking. Just as the vocabularies of two languages never perfectly correspond, so the vocabulary of one’s own language only approximates the “vocabulary” – the concepts – of our thoughts. When I talk to you I am encoding my thoughts, ideas, beliefs into English. This is a process of translation / interpretation which necessarily loses information. When you read my English you attempt to decode the meaning using the phrasebook of your pre-existing understanding, beliefs and expectations. This again is a process of interpretation. Between my ideas and your understanding of them (even supposing I am making sense, possibly a moot point) there is already considerable potential for miscommunication, and we share a common language. It is likely we share a common “culture” too – that of western psychiatry and medicine (like it or not). When working with a client who speaks a different language from you, the interpretation between those languages is variably, and sometimes considerably, prone to information loss or distortion just from a lack of vocabulary equivalence. In addition, you and the client will likely come from different cultures – ‘culture’ meaning in this case not just sociological and anthropological differences but cognitive and conceptual ones. A Chinese person explaining parenting to you in terms of “guan” is presenting a concept alien to Western Europe and not a part of western psychological parenting work. The biggest pitfall of using an interpreter is assuming that you not only understand what the other person is saying, but also what they mean. This pitfall is not confined to interpreting situations.

Interpreters can work in a variety of different ways, the simplest model of which being a notional continuum of verbatim translation to interpretive meaning - or from a word-for-word level through noun-phrase, sentence, paragraph levels to whole document/speech/session levels. The lack of lexical correspondence between two languages renders verbatim interpretation a nonsense, and the other end of the continuum leads to practical difficulties. However the model is useful when considering what you want the interpreter to do.


Conducting a psychological assessment of a psychotic client using an interpreter is arduous. The only time it does not feel that way is if you are not aware of the complications that the interpretation is bringing to the assessment. Interpreters constantly tread a line between ‘telling you what the person said’ and extracting meaning from the utterance and presenting that – and you don’t know which they are doing! Quite often the absence or distortion of meaning is the clinically relevant feature, but it is the task of the interpreter to understand and, quite professionally, they may clarify and clarify with the client until meaning is found in the language when potentially there was none there to start with. With no clear word-for-word translation possible, how can one interpret word salad, for example? Conversely, if the client appears to make no sense how can thought disorder be distinguished from communication difficulties or difficulties of interpretation? Without an interpreter there are still, of course, two interpreting processes going on…


Nascent employees of the Deaf Directorate almost always learn within a matter of days that prelingually and profoundly deaf people with psychotic disorders hear voices. In a service that, upon recruitment, immediately deskills us, fascinating titbits such as this are survival aids. This assertion is rarely questioned, and much has been written about the neuronal activity of deaf hallucinating psychotic people, the phenomenology of deaf psychotic voices, and so on. Perhaps obviously to those outside of this field, the assertion makes no sense. Clinically I have rarely if ever managed to have a (one to one) conversation with a deaf person said to hear voices where the deaf person has been able to describe the pitch, timbre, vocabulary, volume, or location of the voice. Even if there were some way of demonstrating (with a magical functional MRI) that the person really was experiencing an English-speaking female voice of rather low volume, telegraphic prosody, commenting on her clothes from behind her - volume, prosody, the gender of the voice and auditory location are not things that a prelingually profound deaf person could reliably report upon.

Could it be that asking specific questions about such symptoms gives the client (deaf or hearing) a lens through which to evaluate quite probably chaotic and confusing internal experiences? Why do we trust the report of a prelingually profoundly deaf person on the nature of their psychotic experiences when that same client, at the same time, is not able to reliably report upon where they think they are or who they are talking to? Psychotic experiences may well be (who knows?) the cognitive equivalent of dropping all the pages of your thoughts in a puddle and reassembling them, soggy, in the wrong order. The process of encoding those experiences must be even more prone to misrepresentation than non-psychotic mental phenomena. In the context of a clinical interview, deliberately or otherwise asking questions which are even only slightly leading (“Do you sometimes hear a voice that seems to come from nowhere?”) is equivalent to showing a Rorschach blot and asking “Do you see a bat?”

The ABC model of voices suggests the voice is an activating event triggering beliefs about it. Quite possibly the voice is delusional as well. But what’s the difference?

Jim CromwellComment