The first piece of research I ever did was a study of interpreters. I did it on a visit to northern Nigeria as a medical student. I made audiotapes of out-patient consultations between patients who spoke Hausa and doctors who spoke English. I used non-medical staff as interpreters. Once I had made the tapes, I asked some local medical students, all perfectly bilingual, to listen to them and translate all the Hausa statements. I looked at what the patients had actually said and then compared this with what was transmitted by the interpreters.
Not surprisingly, I found all sorts of deviations between the patients’ utterances and what the interpreters conveyed. Some of these deviations were quite legitimate. For example, skilled staff could run through part of a medical history on their own initiative and then report, quite accurately: ‘His urine’s normal’ or ‘He doesn’t have any breathing problems’. On the other hand, I discovered some alarming errors, where interpreters had reported that the patients had said something they had not. Also, the interpreters were pretty selective in what they wanted the doctor to know, and occasionally they even berated patients for what they considered to be vagueness or inconsistencies.
I wrote some guidelines for using interpreters, based on what I found. They went as follows (please excuse the sexist language – it was the 1970s): ‘Greet the patient to establish direct contact. Be seen to be in charge of the interpreter. Assess the interpreter’s English and try to find out how well he speaks the patient’s language. Assess his interests: he may be an anxious relative or an indifferent auxiliary. Give only short sentences for translation, and get the interpreter to explain that the patient must do the same. Make sure everything is translated. Check every answer by asking questions in two or three ways. Finally, use the interpreter to tell the patient everything you would tell him if he could speak your language.’
Forty years later, I would stand by a few of these rubrics, especially the first and last. But I would recant most of the others. Through experience of using interpreters a great deal myself, I have lost my conviction in what one might call the more obsessive, even dictatorial parts of this guidance. Nowadays I believe that the presence of an interpreter, whether in person or by phone link, radically changes the nature of a medical encounter, so that it may be an error to try to force it to resemble something it is not – a one-to-one conversation in a single language. I have even come to think that using interpreters may teach us something important about the nature of all medical encounters and how to conduct them.
I have noticed, for example, that patients speaking in their own language do not like to have their narrative flow interrupted for translation, even when the doctor believes that this needs to happen for diagnostic reasons. To put it at its simplest, it appears that patients want quite literally to be heard, and they may care relatively little whether the main hearer is someone with medical skills or not. I have also noticed that most interpreters, whether they are close relatives of the patient or paid professionals, have personal resources that go beyond the skill of literal, word-for-word, translation. They usually bring to the conversation a wealth of shared cultural associations, and an ability to contextualise utterances that might otherwise be meaningless. Such resources, of course, disprove the notion that ‘pure’ translation actually exists – an idea that no professional literary translator would ever hold.
As I have become aware of these things, I have increasingly tried to sit back and watch with respect as apparently effective consultations unfold between patient and interpreter in spite of – or because of – my passivity. I am content to understand very little for several minutes at a time, if it looks as if the patient is happy with me doing this. I do this particularly with interpreters whom I have used on many occasions before, and know will alert me if anything worries them from a medical point of view. I take the risk of believing that the exchange is serving the purpose that the patient needs, and I try to suppress a wish to intrude my own professional purposes instead.
I suspect that many of my medical colleagues would feel very uncomfortable with this confession. Equally, I have no doubt that many anthropologists and sociologists would approve. Much contemporary social science research shows how doctors are naïve when they imagine that patients tell us their histories only so that we can formulate an accurate diagnosis or recommend a treatment. These are usually our main preoccupations as doctors, but they may not be the patient’s. Illness narratives serve far more urgent and essential purposes. For example, they allow patients, through the very act of speaking, to fashion their memories and sensations into a coherent shape. They provide people with opportunities to assign causation, purpose and direction to their experiences, and to claim moral legitimacy for their own actions. Like all personal narratives, they enable people to describe who they are, to discover who they are becoming, and to define who they wish to be.
Telling a story to another person about one’s experiences is a way of locating the self, perhaps even of constructing it. For such a task, it may be that an interpreter is at least as good a collaborator as a doctor, and may be considerably better. Perhaps we should all have interpreters present when we visit our doctors.