The second set of questions concerns the correspondence of language, both between researcher and respondent and between different respondents. The most critical possibility would be in asking people 'what is sickle cell?' or 'what is beta-thalassaemia'. If they said they did not know, this might be because the condition, or at least its symptoms, was only known to them by a lay term or folk concept. This is theoretically possible, but subjec-tivists do not seem concerned to establish the extent of any intersubjective disagreement (Hindess, 1973).
Genetic conditions mitigate against the development of folk constructs (Rapp, 1988), though these do exist where the prevalence of sickle cell is high (Ebomoyi, 1988). In Britain the more plausible scenario is one where either a child with sickle cell remains undiagnosed, in which case their mortality is likely to be higher (Vichinsky et al, 1988; Powars, 1989), and is unlikely to be given any label, even a folk construct. Or, through contact with the health services a diagnostic label of sickle cell is given, though this may be belated and/or rejected and/or not communicated to the child (Hill, 1994). In developing countries it seems likely that a beta-thalassaemia death would be indistinguishable from other deaths, unless a diagnosis had been made. The author's own research in Britain has utilised several community workers representing Gujarati, Punjabi, Pakistani, Bangladeshi, Chinese, African and Caribbean communities. Neither training days nor debriefing interviews established such lay constructs for the haemoglobin disorders.
More specific breakdowns of intersubjectivity, such as the elderly Gujarati respondent who used the word "thalassaemia major" to evoke the name of the former British Prime Minister in the mistaken but apparent belief that he was being assessed for nationality as part of an immigration check, move us to a third area of concern, namely the different cultural interpretations of the same situation. Statistical risk presumes an abstract mathematical world that may not be shared by those counselled (Rapp, 1988). Multiple choice format schedules may not be understood by all respondents and by the researcher in the same manner. Certainly, in one community survey conducted by the author (Dyson et al, 1993b) a Bengali interviewer apparently interviewed a group of respondents together which at least suggests a conception of knowledge as residing in individuals and to be elicited on a one-to-one basis to be ethnocentric.
Furthermore, the interview situation may have a number of different meanings to respondents, including the intrusion of officialdom; fear of stigma; opportunity to complain about other illnesses as well as an opportunity to gain information (Dyson, 1995). Not all of these constructions threaten an assessment of how much people know about the haemo-globinopathy in question. Moreover, recent work on the co-production of accounts of health (Radley and Billig, 1996) mitigate against accepting the principle of error from an alleged 'true rate' of knowledge.
If the data generated by multiple choice formats are so subject to caveats, and the interview situation and cultural meanings of genetic information are so variable at so many different levels, then is there any purpose in persisting with attempts to assess community awareness of the haemoglobin disorders, when no "real" level of knowledge can be un-problematically stated? Indeed, such surveys of knowledge might be conceived as part of the process of imposing a professional discourse onto communities. But uncovering the ethnographic meanings of the haemoglobinopathies to communities risks revealing lay concepts and strategies to the health professional with no guarantee that information will not be used against lay communities, the more effectively to co-opt them to professional discourses.
It seems that whilst giving information about the haemoglobinopathies is value-laden in that it imposes new invidious reproductive choices onto communities (Rothman, 1994), a general level of community awareness at least creates the necessary, if not the sufficient, conditions for a moral debate. That debate is certainly foreclosed if people, mostly women from minority ethnic groups, are presented with new, devastating, informa tion as isolated individuals at an ante-natal clinic. It is this concern which particularly motivates some realist critics of the new genetics to which we now turn.
Was this article helpful?