Sociology of Health & Illness Vol. 22 No. 6 2000 ISSN 0141±9889, pp. 815±839 Beliefs and accounts of illness. Views from two Cantonese-speaking communities in England Lindsay Prior,1 Pang Lai Chun and See Beng Huat 1 Cardiff School of Social Sciences Abstract This paper examines lay accounts of illness and health gathered ± by means of eight focus groups ± from people living in two Cantonese-speaking communities in England. The authors concentrate on the manner in which Cantonese speakers recruit and mobilise various agents ± such as traditional Chinese medicine (TCM), spirits, demons, food and the weather ± to describe and explain aspects of their bodily and mental wellbeing. As well as providing information on what Cantonese speakers say about such matters, the data are also used to indicate how a concentration on `accounts', rather than on `beliefs', enables sociology to side-step a concern with the subjective and psychological and to focus, instead, on what is publicly available and verifiable. Keywords: Health beliefs, accounts, Chinese On the status of lay belief Until recently, prefacing the word belief with the term `lay' seemed somewhat extravagant. This is mainly because in the social scientific world belief has been commonly regarded as the preserve of the lay public, whilst knowledge and expertise have been associated with the realm of health professionals. Patients hold beliefs about illness, doctors know about them. During the 1990s, however, there appears to have been a major reassessment of the ways in which these various terms are ordinarily combined. Most notably, the words `knowledge' and `expertise' have come to be freely associated with the lay public as much as with professionals, whilst `belief' has tended to be downgraded, and viewed as, somehow, a lesser form of knowledge (see, for example, Popay and Williams 1996). This # Blackwell Publishers Ltd/Editorial Board 2000. Published by Blackwell Publishers, 108 Cowley Road, Oxford OX4 1JF, UK and 350 Main Street, Malden MA 02148, USA. 816 Lindsay Prior, Pang Lai Chun and See Beng Huat switch of emphasis is appropriate, it is claimed, because lay knowledge is as important as scientific knowledge for an understanding of disease and illness, and consequently is deserving of parity of esteem (see also Moffatt et al. 1995, for a similar argument). This contrasts somewhat with earlier standpoints in the sociology and anthropology of health and illness (such as that adopted by, say, Kleinman 1978, 1980, or Eisenberg 1977), which distinguished clearly between lay and professional viewpoints ± a distinction that was, in large part, built around the twin concepts of `disease' (seen as a biomedical reality) and `illness' (viewed as a social interpretation). Indeed, in their overview of the various ways in which `folk' (or lay) health beliefs were interpreted in medical anthropology up to the 1990s, Farmer and Good (1991) point out that it was often the `irrational' and unscientific nature of lay belief that was emphasised. Thus, even in critical Marxist discourse, they claim, `common-sense' and popular thinking about health and illness were viewed as being contaminated by misrepresentation and misconception. `Belief', in ordinary everyday usage, implies trust or faith or even certainty of some kind. In a late modern world where such notions tend to be regarded as heretical, or, at best, will-o'-the wisp, it is not surprising that the word should accumulate negative connotations (see Good 1994). Moreover, in line with a lack of concern with belief per se, authors such as Popay and Williams (1996: 766), Lambert and Rose (1996: 81) and Brown (1995) are also keen to question the notion that scientific medicine holds the primary vantage point from which health and disease can be understood, or that there are clear lines of division between scientific and lay knowledge (Williams et al. 1995: 118). Such a position is, of course, de rigueur in a postmodern ethic that regards all forms of knowledge as situated, particular and provisional (Hollinger 1993) ± and, to that extent, equivalent. In the wake of the broader intellectual climate it is, perhaps, also understandable why lay beliefs and lay knowledge have been re-evaluated in recent empirical work. We see this in studies relating to the new genetics, (for example, Kerr et al. 1998a; 1998b). It has appeared in Arskey's (1998) work on Repetitive Strain Injury, where she has attempted to `blur' the differences `assumed to exist between expert and lay systems of knowledge' (1998: 9). And a similar theme has emerged in Epstein's (1996) study of AIDS activists where he argues that it is often very difficult to determine who is a `layperson' and who is an `expert' (Epstein 1996: 3). In their study of lay beliefs concerning the epidemiology of heart disease Davison et al. (1991: 5) argue that the `traditional lay/scientific dichotomy may well have outlived its usefulness'. In the realm of psychology, Furnham (1998) and Furnham and Kuyken (1991) having examined the content and nature of lay beliefs at length, argue that `there are more similarities than differences' between lay and academic theories ± of, for example, illness and disease (1991: 330). Indeed, the reassessment of lay knowledge has given rise to a # Blackwell Publishers Ltd/Editorial Board 2000 Views on illness and health from two Cantonese-speaking communities 817 new hybrid; the `lay-expert' (Kerr et al. 1998a: 58). The emergence of the latter is predicated on the observation that both lay and professional participants in health care systems have knowledge, and in applying that knowledge they demonstrate expertise. As one would expect, not every empirical researcher of lay beliefs has argued along such postmodernist lines. Thus Richards (1996), for example, is quite ready to maintain a sharp distinction between lay and professional understandings of genetics and inheritance. Similarly, Brown (1992, 1995, 1997), whilst recognising the contribution of lay activists in the realm of `lay epidemiology', implies that lay knowledge, however useful it may be, is of an order that needs to be supplemented with scientific expertise in order to carry political clout. In this paper, we wish to outline an alternative approach to these issues. We do so in the context of a study concerning the health beliefs of Chinese people in England. In the first instance, we argue that what is wrong with a focus on belief is not so much that it denies parity of esteem with expert knowledge, nor that it is out of tune with the democratising tone of the postmodern ethic, but that it implies an inner state of believing. That is to say, a focus on `belief' is suggestive of a psychological state that is somehow locked in individual minds. And although many social researchers refer to belief (as we have shown), we suspect that few would claim to have access to inner psychological states of believing or, indeed, of what is `known'. What social researchers can legitimately lay claim to is `accounts'. That is to say, they can rightly claim access to verbal accounts of what people believe. Such accounts are most frequently obtained from interview data (ours are obtained from focus group interchanges). Accounts are publicly available. They can be checked and re-checked without recourse to a secret inner mind in a way that beliefs cannot be so checked. And they contain claims and narratives, and identify problems, and their solutions are for one and all to see ± so that conjecture about what people inwardly hold to be true or false is unnecessary. Radley and Billig (1996) consider the point crucial and have persuasively suggested that researchers `shift their attention from ``beliefs'' to ``accounts'', in order to analyse what individuals say about health and illness' (1996: 221). The latter is, of course, an argument reminiscent of the one that Gerth and Mills (1953) applied to the concept of motive. They had argued that motives were not buried deep within individual minds, but rather lodged in a collective culture. The rhetoric of `motives' provides a set of socially accepted reasons for acting and as such can be drawn down to account for action in a given circumstance. Such `vocabularies of motive' are therefore socially grounded. The same might be said of health `beliefs'. For they too are lodged in cultural networks and subsequently called upon or recruited to account for observations, events, occurrences. This notion of recruiting or enrolling a `belief' to account for an action, or to provide a justification for not acting, resonates in many respects with recent work in sociological studies of technology. Thus Bijker (1989), in a framework referred to as # Blackwell Publishers Ltd/Editorial Board 2000 818 Lindsay Prior, Pang Lai Chun and See Beng Huat SCOT (the social construction of technology), speaks for example of technological frames. Such frames define what a (technological) problem is and how a solution might be found for it. Different frames identify different problems and are suggestive of different solutions to such problems ± even with respect to what might be considered the same object (Bijker refers to the development of plastic as a suitable example). In the process of identifying and solving problems, networks of people, things and concepts are enrolled and mobilised to support a given position. Tracing the history and social dynamics of the frame then becomes a key task of sociological analysis. The broad lines of this approach, of course, also find parallels in the work of Latour (1999) and other `actor-network' theorists (Law and Hassard 1999). These ideas are of interest to us, then, because they suggest that what ought to be researched in the realm of health beliefs/knowledge is not what is believed or `known', but rather what it is that is enrolled into people's accounts of health and illness. Of equal interest is whether there are social variations in the enrolling and accounting processes, and what kinds of allies and strategies are drawn upon by different parties to advance their claims and counter-claims about the nature, sources and trajectories of specific forms of health and illness. Naturally, the general sociological strategy can be applied to the accounts of medical professionals every bit as much as to those of lay individuals (see, for example, Banks and Prior 2000). Under certain conditions, of course, factions within lay and professional constituencies will form alliances so as to pursue and press claims and arguments about the nature and causes of a specific illness ± as is indicated in the work of Arksey (1998) on the nature of RSI, Brown (1995) on the effects of toxic waste and Epstein (1996) on HIV/AIDS; though the process of recruiting factions is most obvious in areas of medical controversy. In this paper we deal not with medical controversy, but with everyday accounts of health and illness in Cantonese-speaking communities in England. Our task is to map out the kinds of `agents' that are drawn upon by Cantonese-speaking people to account for health and illness; the frames they use to define problems and to seek solutions to such problems; and how the processes of enrolment and recruitment of agents might vary according to the social location of the speaker. As we shall see, the range of agents available to Cantonese-speaking people is large and varied and the techniques of enrolment are complex. Running through all these processes is, of course, the agent commonly referred to as TCM ± which is a complex object in itself. What we wish to indicate herein, then, is that what distinguishes the Cantonese-speaking community from others (both lay and professional) is not so much variations in the `beliefs' that are displayed, but rather differences in the range and nature of objects that are recruited and assembled into accounts of health and illness. Before turning to such matters, however, we first outline the procedures that were used for data collection. # Blackwell Publishers Ltd/Editorial Board 2000 Views on illness and health from two Cantonese-speaking communities 819 Methods of data collection The work on which we report was originally commissioned for the Health Education Authority in London. As far as the latter were concerned, it constituted work preliminary to a survey on the health beliefs of Chinese people in England that was to be executed by what was then known as SCPR (Social and Community Planning Research) during 1998. Our work was initially summarised and reported on during 1997 (Prior et al. 1997), and was used in the planning of the community survey. The key research issues addressed by the focus group work were to determine what kinds of problems the Cantonese-speaking population in England identified as health problems and how they set about dealing with such problems. The survey (to be published during 2000, and which was concerned with Chinese people from various linguistic backgrounds), sought information on the self-identified health problems of respondents, their use of the health service (including the dental services), the use of traditional Chinese medicine, patterns of cigarette consumption, diet and physical activity. According to 1991 Census figures, those who self-identify as `Chinese' comprise only 0.3 per cent of the population in the UK (Coleman and Salt 1997). It is a population that is both geographically dispersed and heterogeneous in its social composition (Cheng 1997). We should also add that it is a diverse population in terms of country of birth and of languages and dialects that are spoken, and one that is poorly researched as compared to other ethnic minority groups (see Parker 1994, Nazroo 1997). Focus groups naturally offer a practical way of zooming in directly on such a hard-toreach population. More importantly, they offer an indispensable means for gaining access to knowledge about publicly-expressed group norms ± as distinct from individual and personal responses that might arise out of the one-to-one interview. The use of focus groups in health research, generally, has been expounded upon in a number of publications (for example, Barbour 1995, Barbour and Kitzinger 1999). Interested readers may draw upon further references in those sources. We are aware, of course, that public (group) and `private' accounts of beliefs about health and illness may differ in the same population according to the context in which people speak (see Radley and Billig 1996). And an interesting parallel in this regard may be drawn with Gilbert and Mulkay's (1984) work on scientist's discourse ± especially in so far as they demonstrate how `talk' about science differs according to the setting in which it is offered. In the light of that, and other studies, it is evident that focus-group data offer an additional dimension to qualitative studies of social life. In the context of this paper, it is also clear that focus groups can provide crucial data on how members of various age, gender and ethnic groups differ in the ways in which they assemble explanatory models of illness and related accounts of healthrelated behaviours. # Blackwell Publishers Ltd/Editorial Board 2000 Northtown Ferrytown Group 1 Females Group 2 Males Group 3 Females Group 4 Males Group 5 Females Group 6 Males Group 7 Females Group 8 Males 19.75 18±21 8 39.5 35±50 0 63.1 62±65 1 32.0 25±39 1 40.0 33±50 4 19.6 19±23 0 32.0 25±36 4 57.6 52±63 1 NEA+ Student Restaurant Nil 7 1 Nil Nil 5 8 Nil Nil Nil Nil 1 4 Nil 1 2 5 2 6 Nil Nil 7 Nil 1 Other Nil 1 Nil 4 6 1 2 Nil 8 6 8 5 11 10 8 8 Mean Age Age Range Number not Married Occupational Locus (Numbers in category) Numbers + Not economically active 820 Lindsay Prior, Pang Lai Chun and See Beng Huat # Blackwell Publishers Ltd/Editorial Board 2000 Table 1 Some Characteristics of Focus Group Participants Views on illness and health from two Cantonese-speaking communities 821 Our focus groups were recruited in two urban areas: one area in the North of England and another in the South. We shall call them Northtown and Ferrytown. At the outset we were interested in recruiting groups of males and females in three age bands. These were roughly of people aged 18±21 years, 25±45 years and 60 years or more. Recruitment was executed by a market research organisation. Respondents were recruited on the basis that they could speak Cantonese and that they were not engaged in health-related occupations (we deliberately wished to exclude `experts'). Once recruited, the researchers met the focus groups in hotels and restaurants close to the participant's place of residence. During the sessions, Chinese tea was served and at the end of the session participants were paid a small sum for expenses incurred. Sessions lasted about 75 minutes. The main characteristics of the focus-group members are summarised in Table 1. The meetings were conducted in Cantonese and to our knowledge this is one of very few pieces of UK health research to be conducted in that language. (Members of the youngest age groups, however, tended to move into English more frequently than we would have wished.) Proceedings were audiotaped and later translated and transcribed (by the second author). Funding did not allow for a second translator, but we do not believe that detracts from the value of our arguments ± which do not hinge on the translation of individual words and phrases. Our opening gambit involved asking participants to react to brief and deliberately vague symptom patterns (cf. Koo 1984: 758) that we presented as individual `cases' of illness. These are summarised in Table 2. Table 2 The Vignettes Case Number Summary Description on Card 1 Woman aged 25. Waking unusually early in the morning. Crying without any apparent reason. Loss of appetite. 2 Three month old baby. Vomiting, diarrhoea, high temperature 3 Woman aged 65. Runny nose, slight temperature. 4 Man aged 45. Dizziness. Headaches. Blurred vision. 5 Female 50. Lumps evident in breast. Otherwise well. 6 Male 50. Lumps evident in upper chest. Otherwise well. We would emphasise, however, that our brief descriptions were used only as stimulants to discussion, and that the focus-group leader posed direct and more telling questions about understandings of health and illness on the # Blackwell Publishers Ltd/Editorial Board 2000 822 Lindsay Prior, Pang Lai Chun and See Beng Huat basis of the response to such examples. We used such cases because we felt that it would be particularly useful to ground any discussion about health in terms of specific examples, rather than abstract principles. To our minds, of course, the symptom patterns were suggestive of identifiable conditions, and abstract or generalised (epidemiological) knowledge could have been brought to bear on the analysis of each case had the respondents wanted. Thus the symptom pattern of the young woman was drawn up with an eye to developing discussions about clinical depression (yõÁ yuÁzheÁng) and `nerves' (sheÂnjingbõÁ ng) in the Chinese community. Cases 5 and 6 are suggestive of a malignant growth (aÂiliuÂ) and so on. Cases were both read out to the participants and presented on a card that was set amidst the tea table. Participants were then asked to discuss three issues. Did the details depicted in the case indicate the presence of a problem that required help in any way? If so, then what kind of help and attention might be appropriate? Where might they best get that help? After about 60 minutes, participants were asked to tell us about `what makes for a healthy life?'. We would draw attention to the fact that vignettes of a similar kind have been successfully used in other studies of psychiatric and physical illness (Lloyd et al. 1998, Hughes 1998). And the findings that we report on below certainly resonate with those derived from the Lloyd et al. (1998) study. Using traditional Chinese medicine (TCM) In professional literature and everyday conversation, TCM is often conceptualised as a unity, as a single object that unites a theoretical system with a coherent body of practice. (In the SCPR survey mentioned above, for example, TCM was translated into visits to `traditional Chinese doctors'). Yet, as Unschuld (1987) has argued, the notion of a coherent Chinese medicine is essentially a western one. For the detailed history of such medicine, he points out, reveals the presence of many conceptual frames ± demonology, cosmology and ancestral ideas included (see also Hanson 1998). Consequently, it is a rather `arbitrary decision' to single out the medicine of correspondence, as Unschuld calls it, to represent TCM. Indeed, Chinese medical systems, like the country, are vast and diverse. Thus, in a recent review of historical studies on the subject, Sivin (1998) has indicated how medical ideas and practices in China varied, significantly, according to locality and region, and the nature of the population (eÂlite versus peasant). Histories of such medicine tend, however, to focus on the texts of elite authors and to ignore health care that was performed by priests and peasants alike. Indeed, it seems likely that the coherence and systematisation of Chinese (as with Ayurvedic) medicines is most likely to exist only in textual forms. In short, the texts confer a misleading sense of unity on practices that were distinctly heterogeneous. What is more, claims Sivin, even to think in terms of a single Chinese medical `theory', is simply erroneous (1998: 743). # Blackwell Publishers Ltd/Editorial Board 2000 Views on illness and health from two Cantonese-speaking communities 823 Such doubts about the relevance of an assumed theoretical system of Chinese medical knowledge to everyday affairs are further reinforced by Farquhar (1987). Speaking specifically of Chinese medicine (she eschews the word traditional), she notes that, in anthropology, `there is a tendency to assemble an internally coherent body of inert knowledge ± a ``system'' ± the ``theoretical foundations'' of which are absolute, idealised and ahistorical. Practice then becomes an instantiation of this ``epistemology'', reproducing knowledge or ``beliefs''' (1987: 1014). One implication of this argument is that anthropologists are sometimes over keen to read sophisticated theory ± derived from texts ± into the basic, everyday health practices of lay people. Indeed, in a different cultural context, Lambert (1992) has noted a tendency in anthropological studies of medical knowledge to ignore the `highly pluralistic and far less clearly systematised' (1992: 1070) schemes operated by lay actors, in favour of an analysis of textually based systems. In studies of Chinese health beliefs, the tendency to read sophisticated medical theory into lay accounts is fairly wide. Thus, discussions elaborating yin/yang dichotomies and `hot'/`cold' distinctions can figure prominently in the analysis of such beliefs (Koo 1984, Gervais and Jovchelovitch 1998a). Yet, and for the reasons stated above, there are good grounds for arguing that where such systems are documented they are more properly viewed as the product of cultural experts (including social scientists) and of other professional commentators, than of lay informants. Understandably, Gervais and Jovchelovitch (1998a), for example, openly acknowledge their debt to expert informants in the field of TCM. Of course, it is not necessary for us to decide on the claims of Farquhar, Sivin and Unschuld in order to conclude that TCM can be viewed as an object (a construction) to be recruited by various parties ± lay and professional (including anthropologists, survey designers and historians) ± for multiple purposes. Our interest is simply in the ways in which TCM may or may not be recruited by Cantonese speakers in England. Before we document the detail relating to our own study, however, it may be as well to mention that even studies of lay health belief in Hong Kong itself (for example, Lee 1997, Koo 1987), would suggest that what has come to be regarded as traditional medical theory plays only a minor role in discussion of illness and its treatment. Thus, for example, Lam et al. (1994) point out that, whilst many Hong Kong Chinese might express preference for Chinese as against Western `tonics' in the domestic treatment of illness, they rarely had any knowledge of what such tonics were for, nor why they might be useful. Further, there is evidence to suggest that, in the Diaspora at least, such factors as age and gender can impact on the issue as to whether TCM is picked up and used in everyday life, far more powerfully than `culture' per se, (see, for example, Anderson et al. 1995). As we shall note, this last claim certainly fits with our own findings. For TCM is not universally accepted in the Chinese community as the preferred or even an acceptable alternative to Western medicine. # Blackwell Publishers Ltd/Editorial Board 2000 824 Lindsay Prior, Pang Lai Chun and See Beng Huat On the use of treatment systems Under what kinds of conditions and, by whom, then, is TCM commonly used? We need to note that three types of treatment system are routinely available to Cantonese-speaking people. First, there is what might be called popular medicine. That is, the medicine that is used and offered within the lay referral system. In our population it is a treatment system that circulates around the use of teas, soups, tablets, herbal preparations and tonics (buÃyaÁo1). Second, there is Western professional medicine as offered by primary care practitioners or hospital doctors. Finally, there is TCM (zhoÁngyi), incorporating in itself a wide range of theories, therapies and practices ± some medicinal, some physical and some supernatural. Chinese people are adept at knowing when to use each of these three systems and they are also adept at judging the efficacy of the different systems (Gervais and Jovchelovitch (1998b)) for specific problems. And in our work we found that there are markedly different approaches to the suitability of the different systems ± approaches that tended to be related to social background (mainly, though not entirely, concerning the age and country of origin of respondents). There was certainly no single and coherent `Chinese' position on the use of TCM. For example, in discussing the nature of the symptom pattern in our Case 3, the men in Group 4 saw a sequence of possibilities: Ho: She may use her past experience to prescribe herself some medicine, either Western or Chinese/ Chun: Most people keep some Chinese tablets (zhoÁng yaÁopiaÁn). They may use these first for a few days, then they may see a Western doctor if the tablets don't work. The discussants in Group 7 saw a similar development, thus: For old people, they should buy the Western tablets first, but some old people would take Chinese Herb tea which is more gentle, such as Gam Mo tea for the flu. Western tablets can get quick results . . . but Chinese herbs take time. Members of all the groups made reference to the quick results available in Western medicine. Chinese medicine is thus regarded as `slow', but, for those who approve of it, it is believed to deal with the root causes of disease (jõ-bõà ng ge-nyuaÂn) at a more fundamental level than does Western medicine. Western medicine is quick and effective, but it only mops up the symptoms of disorder ± only a Chinese doctor can, for example, truly rid a person of the unhealthy influences that cause disease (xieÂ). Here are two Group 5 women discussing the runny nose symptoms of the 65-year-old: # Blackwell Publishers Ltd/Editorial Board 2000 Views on illness and health from two Cantonese-speaking communities 825 Tan: . . .. Western medicine (xiyaÁo) has side effects/ Lau: Chinese medicine (zhoÁngyaÁo) is better, because it is gentle. I think that this woman has wind and fire in the body. Chinese medicine can take the wind and fire out and I believe that only Chinese medicine can cure/ Sen: I agree that Chinese doctors (zhoÁngyi) are best/ Lai: Western medicines make you sleep/ Despite these references to the use of Chinese preparations and TCM, however, there were clear divisions in the community between those who approved of TCM as a mainstay of treatment and those who were reluctant to call upon it. In general, it was older people and people who had been born in China that sang the praises of TCM. Those born in Britain (who also tended to be younger) often had reservations about the use of Chinese medical preparations and TCM. The following respondent from Group 3 (talking about Case 2) laments her son's loss of faith in Chinese tablets. Referring to minor Chinese medical preparations (zhoÁngyaÁo), she states: Look at our generation. We could manage that type of medicine. Even my son's generation took it. It can cure illnesses. It really works, but I never use this for my grandchild. My son does not allow me to give Chinese medicine to my grandchildren. This reference to a generation effect in the use of Chinese preparations was reinforced in the comments of the younger participants. The latter being markedly more sceptical and wary about the use of Chinese medicine as compared to older people. The following extracts taken from Group 4 illustrate the doubts: PLC: Would you take your child to see a Chinese doctor? Mr YING: No. I wouldn't. This is my principle. A negative position reinforced in the following comment from Mr Young (Group 4): I really don't trust Chinese doctors (zhoÁngyi) because they have no official scientific proof. It is very dangerous. Anyone can be a Chinese doctor. I am not against Chinese doctors, but I would never go to one. Recruiting Chinese medicine to deal with illness is not, however, simply a matter of age. It also depends on the display of symptoms and other contingent factors. For example, not one of the respondents suggested that the baby exhibiting signs of tuÁxieÁ (Case 2) should be treated with anything other than Western medicine (xiyaÁo) administered by Western-trained doctors. There were also numerous suggestions that for any form of acute illness, Western medicine would be superior. Chronic conditions, on the # Blackwell Publishers Ltd/Editorial Board 2000 826 Lindsay Prior, Pang Lai Chun and See Beng Huat other hand, could well be susceptible to amelioration from a TCM practitioner ± though it was often noted that such treatments could be very expensive, and, to that extent, economic factors also impinged on the decision whether to recruit TCM or not. In sum, then, the responses of Chinese people to treatment systems were pragmatic and eclectic ± a stance observed by Koo (1987) in her study of Hong Kong respondents. If a treatment (whether it be Western or Chinese) was thought suitable and effective, then it might be used; if not, then people would move on to another treatment system. Though, as we have noted, some respondents would simply `never' use TCM. Others found it `too expensive'. This is not to say, however, that users would have any deep understanding of what various treatments were supposed to achieve. Or, indeed, that users could point toward any theoretical underpinning of their preference for Chinese medicine (again a position observed in relation to Hong Kong users of Chinese tonics by Lam et al. 1994). `Chinese medicine' (zhoÁngyaÁo), therefore, existed as an amorphous object that could be recruited and praised ± more by older people and those born in China than by younger people and those born in Britain ± for various purposes. In some ways it could be recruited as much as an object to underpin identity (`our way of doing things'), as an object for the alleviation of the symptoms of illness, or pain. And, as the material in the following section indicates, `traditional' medicine could also be recruited to explain the origins of illness. On lay aetiology Kleinman (1978) speaks of distinct explanatory models of illness. Such models are among other things, used to account for the onset and progress of illnesses, and to structure the kinds of treatment regimes that are activated to deal with illness. The notion of an explanatory model suggests, of course, the adoption of a coherent and systematic approach (on behalf of both lay and professional parties) to physical and emotional disorders. Yet, as Unschuld (1987) points out, accounts of health and illness, be they professional or lay, are commonly woven out of materials drawn from many sources. In the context of Chinese medicine, Unschuld mentions, for example, demonology as well as Taoism as sources of lay understanding. In our group discussions, a variety of explanatory styles and content certainly emerged. In this section, however, we focus solely on aetiology. Aetiological considerations were not of prime importance to us, but in the way of everyday discourse such factors are often (though not always) linked to an understanding of symptoms. In our study, for example, aetiology tended to be discussed in relation to all the adult cases ± though not to that of the baby. Contrasting schemes of aetiology were certainly evident. In particular, older people and people who had been born in China (these two groups, as # Blackwell Publishers Ltd/Editorial Board 2000 Views on illness and health from two Cantonese-speaking communities 827 we have been keen to indicate, overlapped considerably) were more likely to refer to what Young (1986) has identified as external factors in the causation of disease. The nature of the climate or the seasons, for example, would figure in their accounts. So too would references to spirits and demons. Among younger people, on the other hand, there tended to be fewer references to external factors and far more emphasis on an ethic of personal responsibility. The latter was in accord with the visions of health and healthiness that we shall refer to later. In any event, younger people tended to identify `looking after oneself properly' (baÁoyaÄng) as an important factor in health behaviour. Cutting across this age division, however, were references to nutrition and illness. In Young's schema, nutrition probably figures as internal/physiological type of factor. Nutrition can also be viewed as a cure for illness and so tends to have a dual role. Thus, in relation to an understanding of our case 4 one of the older women in Group 3 stated: In the Chinese way of thinking there may be two parts of the [four] basic elements [earth, fire, water, air] missing. If there is a lack of fire and lack of earth, this person needs good nutrition Yes, somebody takes good wholesome food for a few months, he will improve. Of the externalising factors, the `environment' appeared in many forms. The quality of air was mentioned most frequently (there were, for example, diseases especially associated with dampness (such as, shibõÁng)). Clean versus dirty air, and damp air versus dry air were, consequently, some of the oppositions mentioned. The seasons also figured. Naturally, all such references might be read as pointing toward a nascent theoretical scheme that located disease as a product of inter-relationships between the human body and the wider environment ± though any such scheme was clearly piecemeal and fragmented. For example, although relationships between humanity and the cosmos occasionally emerged from our discussions, they did not do so in any sophisticated manner. Here is Mr Tu (Group 2) speaking of his granddaughter's uncle's mother, who, he says: . . . loses her senses only with the incoming tides. When the tide is high, her madness (dia-nkuaÁng) is strongest. This is not very scientific (ke-xueÂ), but whenever the tides are high, she is mad. The references to tides rather than the lunar cycle is distinctive, and the underlying reliance on astrological influences ± apparent in everyday occidental, as well as oriental thought (Davison et al. 1992) ± is clearly being used to distinguish between folk and expert scientific explanations for changes in affect and behaviour. This same division between what was scientific and what makes sense to `us' was also evident in other discussions. For example, Mr Su (Group 8), thought that the woman referred to in `Case 1' `might be carrying a spell' and continued to say that: # Blackwell Publishers Ltd/Editorial Board 2000 828 Lindsay Prior, Pang Lai Chun and See Beng Huat In Western countries there are no such things. For Asian people maybe there are evil spirits (moÂzhaÁng) moving in to the body. In a similar vein, one of the men in Group 2 stated: In China, medicine is almost magical [. . .]. Chinese believe in magical things like spirits (sheÂndaÁo). Hong Kong has plenty of spirits, they use spirits to cure illnesses. It is the truth. And, Mr Chang (Group 2) spoke of his son (duÂshe-ngzi) when he was ill as a child. In fact, he provided an interesting narrative to account for the illness. Here, however, we concentrate solely on the reference to an evil eye: The monk [daoÁshõÁ ] said that when my wife gave birth, my son saw something evil and got the illness. Unlike these older men, younger people were more `this worldly' in their accounts of illness, though their references tended to be every bit as fragmented and ad hoc as those of older people. Here is an extract from the women in Group 7. They are discussing the man in case 4: Lin: Not enough sleep can give you headaches, dizziness and blurred vision. Kan: Smoking too many cigarettes, drinking a lot of wine and lack of nutrition, or gambling a lot, which leads to lack of sleep and no time to eat. Ling: Some people go gambling for hours on end. They don't eat. They forget to take water. They sit on chairs until they get piles. . . Whilst discussing the details of Case 3, the women of Group 5 pondered on the nature and prevalence of nasal cancer among Cantonese people. The following woman mixes ideas about food and clean air to account for the pattern: Cantonese like to eat stir-fry, deep fat fried and preserved food, such as salted fish. So they get nose cancer easier. Not because of eating too much salted fish, but because of too much stir-fry and deep fat fries. There are too many fumes when you fry food. These extracts are, then, indicative of a mosaic of ideas. Such a mosaic is, of course, what one might expect to emerge from a general discussion about health among non-specialists. Possible causes are recruited from many domains and tailored to suit individual cases. In that vein, it is LeÂvi-Strauss' (1966: 17) concept of the `bricoleur' that comes to mind here. The latter, as someone who accumulates a bag full of tools for organising and arranging things in the world that have no necessary connection. The tool-bag of the # Blackwell Publishers Ltd/Editorial Board 2000 Views on illness and health from two Cantonese-speaking communities 829 bricoleur serves, therefore, as a container composed of whatever kinds of materials lay around and about at any one moment. Both Western and specific Chinese notions ± say, of yin/yang and hot/cold distinctions ± might, in that context, be assembled to account for illness. Though in our case references to detailed Chinese cosmology failed to emerge with any degree of regularity, force or significance. Only the first of our extracts, above, is suggestive of reference to a theorised system. This is not to deny that such theorisations are important to Chinese culture, but it is to suggest that such conceptualisations are, perhaps, much more likely to be recruited by cultural and other experts than by lay generalists. On the whole, then, lay participants seem to extemporise on illness rather than theorise about it, and to draw on any materials that are at hand to account for observations made in the course of everyday life. In that respect the informants reported on here show a number of similarities with the Welsh respondents reported on in the work of Davison et al. (1991, 1992). For example, in both communities common-sense knowledge is used to account for the state of health of a particular man, woman or child. Knowledge of wider social and population patterns may exist in lay consciousness, but the task of the lay commentator is always to explain why misfortune (illness) befell this particular person at this particular time. Indeed, the problem that focus group participants often voiced was that it was difficult to talk about the nature and causes of illness and health because they had so little personal information on our imaginary cases. In other words, participants clearly felt that good explanatory accounts necessarily have to draw on contingent, but detailed personal information rather than on generalised, epidemiological or theoretical knowledge. We argue, then, that explaining the onset and continuation of illness is not so much a matter of `belief', but a matter of constructing plausible accounts ± accounts that usually take the form of narratives. The focus and content of such narratives vary according to the attendant audience. In our experience, for example, the men in Group 2 were more than happy to dwell on narratives that placed emphasis on the role of spells and magic in the onset of illness. Whilst the men in Group 4 avoided such materials and opted instead for what they considered a distanced scientific approach. Nevertheless, in all the groups a reference to personal histories (drawn from their own experiences) formed an important set of exemplars or cases in terms of which the origins of illness could be comprehended. Indeed, personal experience not only helped people to understand what caused an illness, but also what was and what was not illness in the first instance. Framing symptoms We mentioned earlier, Bijker's (1989) use of the notion of a technological frame. Such frames, he claims, both define what a problem is and the likely # Blackwell Publishers Ltd/Editorial Board 2000 830 Lindsay Prior, Pang Lai Chun and See Beng Huat forms of a solution. It is an idea that can be transferred to the study of medical or illness problems and, in that light, it is relevant to ask questions about the kinds of symptoms people include in (and exclude from) a medical frame, as well as what kinds of solutions they seek to devise for dealing with such problems. A world that is seen to hold at least two coherent medical frames (a Western one and a Chinese one) offers potentially fertile ground for the development of opposing views about the nature and sources of illness. Add to that variety of viewpoints a variety of languages and the ground is even richer for such developments. It is not surprising therefore that older Chinese people, in particular, often viewed themselves as being in conflict with the medical professionals that they encountered. Health professionals, didn't listen to what they (as Chinese people) had to say, and Western doctors were generally very difficult to follow. What is more, Western health professionals got simple things like name order confused and failed to treat people as a Chinese doctor would. Participants who raised such issues naturally accepted that many of the difficulties were based on the use of language (see Prior et al. 1997). Differences of language, however, could not alone explain every aspect of reported doctor-patient antagonism. Here, for example, is an older male respondent. He points clearly enough to language differences as a source of difficulty, but hidden within the extract is a reference to something else. It is something to do with conceptualising illness as a state of being that involves more than diseases in autonomous organs: Racism exists. . . . Western doctors are discriminating to us. Whenever they ask a question they know you can't answer it. For example, my doctor asks ``Are you a headache?'' ``Are you a stomach ache?''. It is very quick, rapid, and then he stops. I can't express myself well. Our children are different. . . . This reported focus of the primary care practitioner on the anatomical site of an illness was regarded by a number of the older Chinese people as peculiar. When Chinese doctors came into contact with their patients, it was claimed, they took into account the whole person and their way of life. Western doctors simply asked where the pain was or `where it hurt' and then provided a prescription. The criticism is, of course, familiar even in a Western frame. Rather than focus on different interpretations of `illness', however, what we wish to do here is to draw attention to what Koo has referred to as the `clinical horizon' (1987: 765). Koo used the term in relation to a discussion of the kinds of things that lay (Chinese) people included within the realm of medicine and illness and the kinds of things that they excluded. One of the most obvious and interesting candidates for testing the level of such an horizon is psychiatric illness and it was to such illness that we turned our attention. # Blackwell Publishers Ltd/Editorial Board 2000 Views on illness and health from two Cantonese-speaking communities 831 Psychiatric problems are regarded as potentially stigmatising in most cultures. And Kleinman (1980) clearly indicates that a psychiatric interpretation of symptoms is generally shunned in Chinese culture. Serious mental illness (dia-nkuaÁng) is certainly stigmatised. Above and beyond that, however, it may be the case that what professionals recognise as symptoms of illness (depressed mood, sleeplessness, suicidal feelings and so on), are regarded by the lay public as non-medical problems ± as problems in living perhaps. In that case a `symptom' may well be recognised as a problem, but regarded as lying beyond the realm of medicine. In order to prompt discussion of psychiatric problems we used our Cases 1 and 4. They were intended to draw out any potential references to depression (yõÁ yuÁzheÁng) and `nerves' (sheÂnjingbõÁ ng). Case one involved reference to a female in her early 20s who was displaying symptoms of loss of appetite, sleep disturbance and frequent crying. A number of responses were made to the vignette. First, we noted that description failed to resonate in any of the male groups. It didn't `make sense' they stated. They had never come across such things before. In so far as there might be a problem it was a family problem, or a problem in personal relationships and not a health problem. Perhaps the woman had lost `face' (diu- miaÁnzõÁ ) and so on. In that respect the `clinical horizon' of Chinese males certainly did not seem to stretch so far as to incorporate symptoms of the type outlined. Here is one (male) response from Group 4: I have never seen this situation before. I have no comment on this . . . but [.] if she cries in front of her husband then she is a whinger. I think that this is a family problem . . . When the vignette was offered to the female groups, however, the circumstances seemed to resonate immediately. Here is a response from a woman in one of the Ferrytown groups: The first time I came to England I felt like I had gone to the moon. I knew nothing. I had rented a tiny room, and when I got up I just stared at the four walls. . . I didn't know why I would cry. I had no appetite and would feel tired. I didn't eat for days. This [pointing to the card that described Case One] was my experience. As in the male groups, however, the members of the all-female groups also tended to interpret the apparent symptoms as a product of social problems (especially family problems). It was certainly not regarded as a clear-cut and identifiable health problem. Indeed, even when the word depression was used, it was not at all evident that it was being used in a recognisable clinical sense. For example, one woman in Group 7 stated that the woman in Case 1 was unhappy and `must be depressed', but she did not think that it was `because of illness' (bõÁ ng zheÁng). To that degree, the women in our groups seemed as reluctant as men to site such symptoms in a clinical frame. # Blackwell Publishers Ltd/Editorial Board 2000 832 Lindsay Prior, Pang Lai Chun and See Beng Huat Here is Ms Ling. Group 7: Or this person has some kind of disease she doesn't know about. Or there are some external influences. Or maybe she has no illness but family events affect her. If she can solve her problem she won't act so abnormally . . . Just sending her to a psychiatrist won't solve her problems. If you understand her problems as being caused by family, husband, children, then [you may be able to do something]. It is also worth noting that the symptom pattern of Case 4 ± which we also intended to be suggestive of psychological stress ± was also excluded from clinical considerations (see, for example, the extract on page 828). And it was certainly not considered as being suggestive of psychological problems. Overall, then, it was only in the female groups that the possibility of such things as psychiatric illness and, specifically, depression, were brought into the discussion at any level. So our results would suggest that an understanding of symptom patterns is clearly related to social position (a gender-structured position in this instance). The significance of a speaker's social location also emerged in other features of our discussions. For example, reactions to the `lumps' in vignettes five and six showed a marked gender differentiation ± women suggesting that some form of preventive action be taken and men being more likely to dismiss the notion that such signs ought to be taken seriously. Despite that, of course, we should be aware that Chinese people evidence very low rates of usage of primary health care and screening services generally. Thus, Smaje and Le Grand (1997), for example, using General Household Survey data, indicate that Chinese people in Britain exhibit particularly low rates of consultation with general practitioners and have the lowest rates of out-patient consultations for any ethnic group, including whites. Similar patterns of use have also been noted among Chinese people in other Western countries ± see, for example, Lam (1994). Consonant with these observations are others to the effect that Chinese women exhibit very low rates of breast self-examination and have comparatively low rates of uptake on cervical screening programmes (Watt et al. 1993). How these low rates of consultation and usage may be accounted for is problematic. With respect to psychiatric problems, however, it could be that it is not simply a matter of shame and `face' that is involved, but a genuine failure to recruit `emotional' symptoms into a clinical frame. For Chinese people ± as with most Europeans ± think of illness only in terms of bodily pain and bodily symptoms. On being healthy (JiaÁn Ka-ng) As was indicated above, questions about what make for a healthy life were left until the end of the focus group sessions. We had originally structured # Blackwell Publishers Ltd/Editorial Board 2000 Views on illness and health from two Cantonese-speaking communities 833 the focus group membership so as to draw out what we assumed might be contrasting visions of health and illness in the Chinese community (say, between male and female, young and old, British born and Hong Kong born). As far as illness (bõÁ ng zheÁng) was concerned we certainly saw some social variations. It was, however, something of a surprise to discover on analysis that, as far as images of health were concerned, there seemed to be a remarkable degree of agreement about how a healthy life was to be achieved. The similarity rested in the appeal to `happiness' as a source of health. A simple content analysis of the translated transcripts, for example, revealed more references to `happiness' (and unhappiness) in the context of talk about health, than any other factor. References to happiness were particularly marked among the female groups. Thus, the latter displayed an average of 14 citations to `happiness' in the context of discussions on health, compared to an average of only five citations for the male groups ± though none of the males in Group 2 made any reference to the term. Here is a selection of straightforward responses to our request for a discussion about the kinds of factors that might be associated with a healthy life: Hong (Group 1): You've got to be happy (kuaÁileÁ) Yau (Group 3): Having a peaceful feeling, happiness (xingfuÂ), satisfaction, these can make you physically and mentally healthy. Mr Chung (Group 4): If your heart is happy then you are healthy Ling (Group 5): If you are happy (kuaÁileÁ), have a happy family, have no heavy pressure, then you will be in good health. Group 6. [With an instant response to the question.]. PLC: I want to know what you think makes people healthy? KC: Being happy. Kim (Group 7): The number one thing is being happy (kuaÁileÁ). This focus on happiness and inner contentment implied that it did not really matter how you behaved in relation to what professionals might consider unhealthy activities. If smoking cigarettes made one happy then it was acceptable to smoke. If eating fatty foods led to contentment then that was acceptable also ± all in the name of happiness. Some participants, of course, pointed toward other factors as being associated with health (exercise and fresh foods were also mentioned), but the overwhelming concern was with this state of inner contentment. Naturally, a full analysis of the concept of health in the Chinese community would require a much more detailed investigation than is reported on here. And were such investigations to be conducted we might indeed find a variety of conceptualisations of the type uncovered by, say, Blaxter (1990) or Williams (1990) in other populations. The latter two authors point out how health can incorporate notions of # Blackwell Publishers Ltd/Editorial Board 2000 834 Lindsay Prior, Pang Lai Chun and See Beng Huat physical functioning, fitness, vitality and moral worth as well as a sense of wellbeing or balance. And in order to indicate the presence of some alternative ideas about health we offer an extract from an interchange between two men in Group 8: [PLC]: My final question. What do you think brings mental and physical health? Mr Su: If you practise Tai Chi or Kung Fu every morning, because in the morning the air is fresh and/ Mr Mac interjects: /I don't think Kung Fu is very important. The main reason is if you have a happy family (kuaÁileÁ jiatõ ng). . . . Mr Su: /If we want to keep young, physically and mentally we really need exercise and fresh air. . . Of course, if your children do not rebel a lot and you can get a happy family . . . To which Mr Mac responds: If you are happy, then you will be healthy. If you can laugh and are joyful and peaceful, you are especially healthy and fresh. If you are not happy, it does not matter how good you are in Tai Chi. By extending reference to family life, (a factor that Kleinman (1980: 147) also highlights as being of considerable significance in Taiwanese health beliefs), this extract is possibly suggestive of a much broader conceptualisation of health than that which simply equates health with happiness. Indeed, on re-reading the last dialogue, we can also catch a glimpse of a notion of `health' that extends way beyond the physical body and into the social body. For our purposes, however, we merely note that of all the entities that could have been enrolled to account for health, it was an inner state of balance or contentment ± a mental state ± that proved to be the most popular candidate. In that sense, we can note that whilst there were a wide variety of agents (nutrition, the environment, demons and spirits) recruited so as to account for illness (bõÁ ng zheÁng), the range of agents recruited to account for health (jiaÁn kha-ng) were much more limited. More interestingly, perhaps, we can also see how reference to `belief' and other mental states such as feelings of being happy, are mobilised into the public process of accounting. In other words, rather than forming the basis for what is said (and possibly done), references to inner mental states (including beliefs) are used as building blocks for accounts, (see also, the extract on p. 828). Conclusion We opened this paper by making reference to a debate concerning the status of lay belief vis aÁ vis professional knowledge. We noted and documented an # Blackwell Publishers Ltd/Editorial Board 2000 Views on illness and health from two Cantonese-speaking communities 835 emergent argument about the equivalence (or otherwise) of lay belief and professional knowledge. One of the claims incorporated into that argument is that reference to `belief' downgrades the value of lay interpretation as against expert interpretation. Whether that is indeed so, it is not necessary for us to judge, for we have argued that there are other and better, reasons for side-stepping a concern with belief (or, indeed, with what is `known'). One such reason concerns the fact that a focus on belief is markedly psychologistic in emphasis. It concentrates attention on what is supposedly within the minds of individual subjects and thus rests on evidence that is unobtainable. What is obtainable is an account (Radley and Billig 1996). Accounts can be gathered from face-to-face interviews, from naturally occurring talk, or from focus-group interchanges. In each instance, of course, the accounts in question might differ one from the other according to the context in which they are offered ± as is apparently the case with expert scientific discourse (Gilbert and Mulkay 1984). As far as focus-group accounts are concerned, it is clear that what people bring forward into the group arena are claims that they consider to be publicly acceptable. They reflect what people think the `community line' is. As such the method offers another dimension of qualitative data for social scientific analysis. And, as we hope to have shown, they are certainly useful for demonstrating contrasts in the accounts of different social groups (male/female; young and old). Our attention has, of course, been on what the accounts make reference to; on what is recruited and assembled within them; on what they connect and disconnect; and on how and, by whom, specific items and things are mobilised. As we have seen, many of our accounts recruited TCM and Western medicine as entities to be mobilised. Though detailed consideration as to exactly how TCM was assembled through text, talk and interaction was a problem that lay beyond the boundaries of this paper. Nevertheless, it was clear that the assembly process melded together various agents (materia medica, experts, concepts and theories) and that the mobilisation of TCM was triggered only under certain circumstances ± for certain kinds of symptoms and people (vomiting, diarrhoeal babies, for example, were excluded) ± and by particular segments of the community. Yet, whether it be zhoÁngyaÁu or xiyaÁo that is sought and used, the conditions that are excluded from the clinical frame of a community are no less interesting than those that are included. In most cultures, of course, wounds and body pains are recognised as things to be included. When it comes to psychological symptoms, on the other hand, opinions often alter. Thus, tears, sleeplessness, irritability and depressed mood and so forth, are often viewed as mere `problems in living', or social problems that lay beyond medical help. That certainly seems to be the case in the Chinese community. Thus, few people seemed to think that the loss of appetite, crying and unusually early morning waking of the young woman was indicative of a medical problem. Even the somatic symptoms of Case 4 only just managed # Blackwell Publishers Ltd/Editorial Board 2000 836 Lindsay Prior, Pang Lai Chun and See Beng Huat to find inclusion in a clinical frame (though not a psychiatric one). And it is, of course, in the interstices between alternative clinical frames that what are referred to as cultural syndromes often appear. So what is but an exaggerated social or personal problem for one group, is regarded as an illness (or even a disease) by another. Texts on Chinese culture highlight a number of such syndromes (such as koro and xie bõÂng), though we have not sought to list them, as they were not mentioned in our focus-group discussions. (For further detail, see Prior et al. 1997.) Naturally, mapping the clinical boundaries of a population such as the one being discussed, could well throw considerable light on patterns and rates of primary care consultations in the community. Our major argument, however, is that what is contained within such boundaries is not to be found by examining people's beliefs ± nor of what they inwardly `know' ± but is a matter of examining their accounts. Accounts link `things', concepts and practices together in a seamless web. In that vein sociological health research becomes a matter not so much of dwelling on subjectivity and the inter-subjective, but of examining issues of enrolment, assembly, and mobilisation ± in short, what Latour (1999) has referred to as patterns of `interobjectivity'. Address for correspondence: Lindsay Prior, Cardiff School of Social Sciences, Glamorgan Building, King Edward VII Avenue, Cardiff CF10 3WT e-mail: PriorL@Cardiff.ac.uk Note 1 Chinese words in brackets are pinyin equivalents of Cantonese terms used in the discussions. See, J. DeFrancis (ed) ABC Chinese-English Dictionary. 1996. London: Curzon Press. References Anderson, J.M., Wiggins, S., Rajwani, R., Holbrook, C., Blue, C. and Ng, M. (1995) Living with a chronic illness: Chinese-Canadian and Euro-Canadian women with diabetes, Social Science and Medicine, 41, 2, 181±95. Arksey, H. (1998) RSI and the Experts. The Construction of Medical Knowledge. London: UCL Press. Banks, J. and Prior, L. (2000) Doing things with illness. The micro-politics of the CFS clinic, Social Science and Medicine (forthcoming). Barbour, R.S. (1995) Using focus groups in general practice research, Family Practice, 12, 3, 328±34. Barbour, R.S. and Kitzinger, J. (eds) (1999) Developing Focus Group Research. London: Sage. Bijker, W.E. (1989) The social construction of Bakelite: toward a theory of invention. In Bijker, W.E., Hughes, T.P. and Pinch, T. (eds) The Social Construction of Technological Systems. Cambridge, MA: MIT Press. # Blackwell Publishers Ltd/Editorial Board 2000 Views on illness and health from two Cantonese-speaking communities 837 Blaxter, M. (1990) Health and Lifestyles. London: Tavistock. Brown, P. (1992) Popular epidemiology and toxic-waste contamination: lay and professional ways of knowing, Journal of Health and Social Behaviour, 33, 267±81. Brown, P. (1995) Popular epidemiology, toxic waste and social movements. In Gabe, J. (ed) Health, Medicine, and Risk. Sociological Approaches. Oxford: Blackwell. Brown, P. (1997) Popular epidemiology revisited, Current Sociology, 45, 3, 137±56. Cheng, Y. (1997) The Chinese: upwardly mobile. In Peach, C. (ed) The Ethnic Minority Populations of Great Britain: Ethnicity in the 1991 Census Series. Volume 2. London: HMSO. Coleman, D. and Salt, J. (eds) (1997) Ethnicity and the 1991 Census. Volume 1 Demographic Characteristics of the Ethnic Minority Populations. London: HMSO. Davison, C., Smith, G.D. and Frankel, S. (1991) Lay epidemiology and the prevention paradox: the implications of coronary candidacy for health education, Sociology of Health and Illness, 13, 1, 1±19. Davison, C., Smith, G.D. and Frankel, S. (1992) The limits of lifestyle: re-assessing `fatalism' in the popular culture of illness prevention, Social Science and Medicine, 34, 6, 675±85. Eisenberg, L. (1977) Disease and illness. Distinctions between professional and popular ideas of sickness, Culture, Medicine and Psychiatry, 1, 9±23. Epstein, S. (1996) Impure Science. AIDS, Activism, and the Politics of Knowledge. Berkeley: University of California Press. Farmer, P. and Good, B. (1991) Illness representations in medical anthropology: a critical review and a case study of the representation of AIDS in Haiti. In Skelton, J.A. and Croyle, R. (eds) Mental Representation in Health and Illness. N.Y.: Springer Verlag. Farquhar, J. (1987) Problems of knowledge in contemporary Chinese medical discourse, Social Science and Medicine, 24, 12, 1013±21. Furnham, A.F. (1988) Lay Theories: Everyday Understanding of Problems in the Social Sciences. Oxford: Pergamon. Furnham, A. and Kuyken, W. (1991) Lay theories of depression, Journal of Social Behaviour and Personality, 6, 2, 329±42. Gerth, H. and Mills, C.W. (1953) Character and Social Structure. New York: Harcourt Brace. Gervais, M-C. and Jovchelovitch, S. (1998a) The Health Beliefs of the Chinese Community in England. London: HEA. Gervais, M-C. and Jovchelovitch, S. (1998b) Health and identity: the case of the Chinese community in England, Social Science Information, 37, 4, 709±29. Gilbert, G.N. and Mulkay, M. (1984) Opening Pandora's Box. A Sociological Analysis of Scientists' Discourse. Cambridge: Cambridge University Press. Good, B.J. (1994) Medicine, Rationality and Experience. An Anthropological Perspective. Cambridge: Cambridge University Press. Hanson, M. (1998) Robust northerners and delicate southerners: the nineteenthcentury invention of a southern medical tradition, Positions. East Asia Cultures Critique, 6, 3, 515±50. Hollinger, D.A. (1993) How wide the circle of the `we'? American intellectuals and the problem of the ethnos since World War II, American Historical Review, April, 317±37. # Blackwell Publishers Ltd/Editorial Board 2000 838 Lindsay Prior, Pang Lai Chun and See Beng Huat Hughes, R. (1998) Considering the vignette technique and its application to the study of drug injecting and HIV risk and safer behaviour, Sociology of Health and Illness, 20, 3, 381±400. Kerr, A., Cunnigham-Burley, S. and Amos, A. (1998a) The new genetics and health: mobilizing lay expertise, Public Understanding of Science, 7, 41±60. Kerr, A., Cunnigham-Burley, S. and Amos, A. (1998b) Drawing the line: an analysis of lay people's discussions about the new genetics, Public Understanding of Science, 7, 113±33. Kleinman, A. (1978) Concepts and a model for the comparison of medical systems as cultural systems, Social Science and Medicine, 12, 85±93. Kleinman, A. (1980) Patients and Healers in the Context of Culture. Los Angeles: University of California Press. Koo, L.C. (1984) The use of food to treat and prevent disease in Chinese culture, Social Science and Medicine, 18, 9, 757±66. Koo, L.C. (1987) Concepts of disease causation, treatment and prevention among Hong Kong Chinese, Social Science and Medicine, 25, 4, 405±17. Lam, C.L.K., Catarivas, M.G., Munro, C. and Lauder, I.J. (1994) Self-medication among Hong Kong Chinese, Social Science of Medicine, 39, 12, 1641±47. Lam, L. (1994) Self-assessment of health status of aged Chinese-Canadians, Journal of Asian and African Studies, 29, 1±2, 77±90. Lambert, H. (1992) The cultural logic of Indian medicine: prognosis and aetiology in Rajasthani popular therapeutics, Social Science and Medicine, 34, 10, 1069±76. Lambert, H. and Rose, H. (1996) Disembodied knowledge? Making sense of medical science. In Irwin, A. and Wynne, B. (eds) Misunderstanding Science? The Public Reconstruction of Science and Technology. Cambridge: Cambridge University Press. Latour, B. (1999) On recalling ANT. In Law, J. and Hassard, J. (eds) Actor Network Theory and After. Oxford: Blackwell. Law, J. and Hassard, J. (eds) (1999) Actor Network Theory and After. Sociological Review Monograph. Oxford: Blackwell. Lee, S. (1997) How lay is lay? Chinese students' perceptions of anorexia nervosa in Hong Kong, Social Science and Medicine, 44, 4, 491±502. LeÂvi-Strauss, C. (1966) The Savage Mind. London: Weidenfeld and Nicolson. Lloyd, K.R., Jacob, K.S., Patel, V. et al. (1998) The development of the short explanatory model interview (SEMI) and its use among primary-care attenders with common mental disorders, Psychological Medicine, 28, 1231±37. Moffatt, S., Phillimore, P., Bhopal, R. and Foy, C. (1995) `If this is what it's doing to our washing, what is it doing to our lungs?' Industrial pollution and public understanding in North-East England, Social Science and Medicine, 41, 6, 883±91. Nazroo, J.Y. (1997) The Health of Britain's Ethnic Minorities. Findings from a National Survey. London: Policy Studies Institute. Parker, D. (1994) Chinese in Britain: Annotated Bibliography and Research Resources. University of Warwick. CRER. Popay, J. and Williams, G. (1996) Public health research and lay knowledge, Social Science and Medicine, 42, 5, 759±68. Prior, L., See, B.H., Pang, L.C. and Bloor, M. (1997) The Health Needs and Health Promotion Issues Relevant to the Chinese Community in England. Cardiff: SOCAS, Cardiff University. # Blackwell Publishers Ltd/Editorial Board 2000 Views on illness and health from two Cantonese-speaking communities 839 Radley, A. and Billig, M. (1996) Accounts of health and illness, dilemmas and representations, Sociology of Health and Illness, 18, 2, 220±40. Richards, M. (1996) Lay and professional knowledge of genetics and inheritance, Public Understanding of Science, 5, 217±30. Sivin, N. (1998) The history of Chinese medicine: now and anon, Positions. East Asia Cultures Critique, 6, 3, 731±62. Smaje, C. and Le Grand, J. (1997) Ethnicity, equity and use of health services in the British NHS, Social Science and Medicine, 45, 3, 485±96. Unschuld, P. (1987) Traditional Chinese medicine: some historical and epistemological reflections, Social Science and Medicine, 24, 12, 1023±29. Watt, I.S., Howell, D. and Lo, L. (1993) The health care experience and health behaviour of the Chinese: a survey based in Hull, Journal of Public Health Medicine, 15, 2, 129±36. Williams, G., Popay, J. and Bissell, P. (1995) Public health risks in the material world: barriers to social movements in health. In Gabe, J. (ed) Health, Medicine, and Risk. Sociological Approaches. Oxford: Blackwell. Williams, R. (1990) A Protestant Legacy. Attitudes to Death and Illness among Older Aberdonians. Oxford: Clarendon. Young, A. (1986) Internalising and externalising medical belief systems: an Ethiopian example. In Currer, C. and Stacey, M. (eds) Concepts of Health, Illness and Disease. Leamington Spa: Berg. # Blackwell Publishers Ltd/Editorial Board 2000