Personal Experience of Illness

Arthur Kleinman & Don Seeman. Handbook of Social Studies in Health and Medicine. Editor: Gary L Albrecht, Ray Fitzpatrick, Susan C Scrimshaw. Sage Publications, 2000.


They need to understand that we don’t live in a health clinic’ These are the words of an Ethiopian-Israeli political activist, protesting his country’s public health policy with respect to the prevention of HIV infection. In particular, he objected to the routinized (but unpublicized) destruction of blood donations made by Ethiopian immigrants, which he believed had stigmatized his community in a way that would have important social and health consequences (Seeman 1997). Indeed, the two can hardly be separated. As experience in the United States and elsewhere has shown, serious consequences can derive from depicting an ethnically defined segment of the population primarily as a vector of risk for infectious disease without an equal degree of concern for their own pressing health needs, or for the political and economic conditions that help to put them at increased risk.

In this case, bureaucratic taxonomies that kept particular groups of potential immigrants languishing in Addis Ababa transit camps for 5 years or more can be correlated with high rates of HIV infection, which were then deployed in arguments for the further restriction of immigration. For the ill and their families, this contributes to an already heavy burden of ostracism within their own ethnic community as well as the general society. The experience of illness becomes overlaid with concern for the collective authenticity of immigrants as citizens and their perceived rebuff by medical institutions. Haitians demonstrating over blood bank policies in New York (Farmer 1992) and Ethiopians demonstrating over blood bank policies in Jerusalem represent important local variations on a set of themes that appear increasingly international in scope, defined by a language of technical efficiency and cost effectiveness, which may leave little room for the concern with experience that we are advocating in this chapter.

The accusation that public health experts and government officials had acted as if the world they inhabited was one devoid of social meanings and free of the structural violence which systematically exposes some groups to risk while denying social responsibility for that risk represents a simplified diagnosis of a specific local problem, but it is a problem that can be heard echoing through many other settings. Whether the specific illness complaint is one of chronic pain whose contours are shaped by the traumatic political upheavals of revolutionary China (Kleinman and Kleinman 1994), or one of the ‘emerging infectious diseases’ whose paths of devastation can be shown to converge with those of impoverishment and international economic exploitation (Farmer 1996), the message that emerges from careful ethnographic research is that illness has a determinative social context and course (see Kleinman et al. 1995). ‘We don’t live in a health clinic’ is a reminder that the social course of illness and the personal experience of illness are analytically inseparable.

Those whose lives are most directly touched by illness—the sick, their families, and members of their social networks (including in some cases their immediate care providers)—can often be heard resisting the reductionistic categories into which their experience is sometimes parsed by both medical professionals and social scientists. The aim of this chapter, therefore, is to suggest a more experientially valid and morally engaged mode of analysis. A useful conception of illness experience must bear sufficient theoretical weight to allow for the sustained analysis of its moral and social bases. It should also speak to the need for deepening both clinical care and research paradigms. The ‘personal experience of illness’ necessarily connects with a broad world of theoretical and practical concern.

Consider the anger of Mavis Williams, mother of a 23-year-old son in the United States who suffers from muscular dystrophy. When she was asked to fill out a questionnaire concerning the effect of her son’s illness on her family, she responded instead with rage towards the investigator: There is a little voice in me which, if I knew you better, would scream at you: Doctor, [my son’s illness] has murdered this family! There is no stability; we can’t work it through … Before Andrew’s disaster we were like everyone else: some days good, some bad … Now we are burning up. I sometimes think we are all dying, not just Andy. Even my parents and brothers and sisters have been more than “affected”

She proceeds to tell the tale of her family; of a husband who collapsed under the burden of his son’s sickness, began to drink, and then disappeared; of two other children whose lives have been framed by the twin specters of guilt (‘because they are normal’) and intense (yet inexpressible) anger towards their mother, who has been absorbed with Andrew’s care. As for Mavis Williams, she is morally and physically spent, and blames herself for much of her family’s demise. ‘You look around you, look!’ she pleads. ‘This, what you see, this tomb, our family’s tomb.’

This is only a brief vignette of a story that has been told in greater detail elsewhere (Kleinman 1988: 183-4), but it confirms an important presupposition of our analysis. The experience of illness is not bounded by the bodies or consciousness of those who are ill. It reaches out to encompass a household, a family, or a social network. It reaches deep into the inner world of patients, yet is decidedly transpersonal. Where is the suffering located in the case of Mavis Williams and her son? In Andrew, dying of a progressively debilitating disorder? In his mother? His siblings? Obviously, the illness experience affects each of them, not in isolation, not always in the same way, but in relation to one another. A questionnaire can ask about symptoms and evaluate disability, but it cannot get at the qualities of experience that are so deeply at stake in this family tragedy.

It is the irreducible ‘sociosomatic’ quality of illness experience, rooted in the infrapolitics of everyday life as well as the more encompassing political, economic, and cultural realities of the societies we live in, which should be of deep concern to both social scientists and medical professionals. When we describe ‘illness experience,’ therefore, we mean something other than the stories of individuals, robbed of social context, cultural mooring, or personal agency and exigency. Illness experience as we are using that term here needs to be recognized for something that is not necessarily transparent to the categories of traditional biomedicine or social science, but more responsive to the contradiction, indeterminacy, and moral weightiness—in the sense of something being desperately at stake—which opens out to us from our encounter with Mavis Williams and her family. For a family such as hers, or for an immigrant group protesting the ‘rational and necessary’ measures taken by public health experts, the experience of illness takes on a wider significance than the view from the clinic might at first suggest.

What Do We Mean by Illness?

One early contribution of medical anthropology, which, with some revision, remains crucial to our discussion of illness experience in this chapter, is the analytic distinction between illness and disease (Kleinman 1980: 72-80). Stated simply, ‘disease’ has been defined in medical anthropology as the practioner’s construction of patient complaints in the technical terminology of a particular healing system. Thus, for biomedicine, ‘disease’ refers to ‘abnormalities in the structure and/or function of organs and organ systems; pathological states whether or not they are culturally recognized; the arena of the biomedical model’ (Young 1982: 264). This is not to say that other professional and folk healing systems (Aryuvedic, traditional Chinese, etc.) do not maintain their own technical concepts of disease, although the term has been less widely used in this context.

‘Illness,’ on the other hand, has been deployed as an ‘experience-near’ category, within which room can be found for the culturally patterned social and personal elements of sickness, which are often excluded as a matter of course from the technically constrained discourse on disease. As Allan Young wrote in his 1982 review article, ‘Anthropologies of Illness and of Sickness,’ ‘illness refers to a person’s perceptions and experiences of certain socially disvalued states, including, but not limited to, disease’ (Young 1982: 265). In this conception, ‘muscular dystrophy’ would be considered by physicians of biomedicine a disease of the central nervous system. Its powerful and destructive ripple effect on the life-world of the Williams family, and on the bodily and emotional health of its members, constitutes the corresponding illness that is often excluded from consideration.

Part of the reason that the dichotomy between illness and disease was so fruitful for medical anthropology—but also perhaps limiting in certain ways—was that it mirrored a defining feature of contemporary biomedicine; namely, the self-avowed reductionism which is typically invoked in the name of instrumental rationality and, with growing urgency today, economic efficiency as well. A particularly striking example, as Byron Good has shown, is the systematic training of American medical students to locate the essence of disease not in the interpersonal context of social and political life, nor even within the contingent life trajectories of particular individuals, but in the ostensibly irreducible cellular and molecular processes which are the subject of increasing technical manipulation (Good 1994). This is a powerful local knowledge feature of contemporary biomedicine (cf. Geertz 1983), and we should not gainsay its contribution to the tremendous international prestige and efficacy of some biomedical interventions.

On the other hand, anthropologists and others have been painfully aware of what can get left out of accounts in this reductionist tradition, which pervades not only the training of medical students, but clinical practice and research as well. The limits of reductionist models are evident in at least three areas, each of which provides a special challenge to medical and social science paradigms, which we hope the study of illness experience can help to address. It is worth remembering, however, that while the distinction between illness and disease described above was originally conceived as an alternative to narrow and exlusionary biomedical accounts, it has not been useful to an equal degree in addressing each of these three areas of concern.

  1. Reductionist accounts may contribute to a regime of knowledge in which the social roots and distribution of disease as well as of resources for care are misrecognized. This occurs when disease is treated as a property of individuals outside of meaningful social and political context. It has been argued convincingly, for instance, that poverty is themajor worldwide risk factor today for mortality due to infectious disease (Farmer 1996). However, contemporary social science and medical accounts alike often focus on pathologies of individual choice and ‘risk behavior’ as if these were self-evident and untheor-ized constructs. Reference to the powerful and historically conditioned constraints on choice in precisely those settings where the burden of disease has been greatest tends to be avoided. Few accounts, for instance, note that poor women in many societies are at special risk for sexually transmitted disease. This is so, not because of their pathological inability to make appropriate behavioral choices, but because existing structures of power and economic inequality render them incapable of refusing the sexual demands of male partners, including their husbands (Farmer et al. 1996).

The perception of injustice, so often a component of illness experience, accurately reflects the structual violence that contributes to differential risk for sickness in many settings. Yet, this feature of social reality is typically absent from biomedical and public health narratives, and it is frequently (although by no means as systematically) absent from alternative professional and folk accounts of disease as well (Leslie and Young 1992). To the extent that we rely on stylized cultural narratives of illness therefore, there is a risk that anthropological accounts will parallel those of biomedicine in this regard. Investigation of the illness experience should always lead back to a consideration of the shifting political, historical, and physiological realities in which it is rooted. The analytic distinction between illness and disease ought not preclude an investigation of their subtle interlinkages.

  1. Reductionist accounts may overlook or exacerbate the disruptive and morally ramified effects of illness on the life-worlds of sufferers. This is one of the problems that the illness model in medical anthropology was explicitly designed to ameliorate. Indeed, it is the contingent depiction of suffering in particular life-worlds that is most often identified with this approach. Not confronting the domain of illness can lead to a serious dehumanization of medical care. A physician who does not attend to the demoralizing effects of life with chronic pain, for example, its ability to drown out those qualities that make life seem worthwhile, and its tendency to raise suspicions of malingering or insincerity by the sufferers, will not adequately respond to the sufferer’s needs (see Good et al. 1992). Worse, the physician may be drawn subtely into the chorus of doubting and frustrated voices which add to sufferers’ misery (Scarry 1985: 4). Attending to the illness, and not just the disease, means taking the socially embedded person seriously. The late philosopher Emmanuel Levinas (1988) suggested that the call of medicine is the call to respond to sufferers’ ‘originary cry for help,’ which includes, but is not limited to, the biomedically defined ‘presenting symptoms’ alone. Biomedical and other cultural idioms enter the stream of illness experience in complex and unpredictable ways, and should be approached not abstractly, but rather phenomenologically, as they appear from within and help to shape that lived context.
  2. Medical anthropology, perhaps more than any other discipline, has shown that biomedicine’s striking (and perhaps unique) attempt to disengage healing from the realm of moral meaning may allow cultural realities, which have great bearing on the course of illness and disease, to be made peripheral to its concern(Kirmayer 1992; Kleinman 1980, 1988). There is by now an extensive literature on the ways in which illness is embedded in particular local systems of meaning and discourse (e.g., Brodwin 1996; Good 1976; Rubel et al. 1984; Taylor 1992). In one study, Karen Pliskin (1987) showed that doctors’ inability to come to terms with the somatized rhetoric of complaint among Iranian immigrant patients led to frustration on the part of doctors and misdiagnosis of some patients as mentally ill.

In the field of mental health, cultural specificity has even attained a level of formal recognition, through the notion of ‘culture bound syndromes,’ which has been adopted within the diagnostic canons of DSM-IIIR and DSM-IV. While some authors have gone so far as to collapse the distinction between illness and disease completely in this setting, claiming that all mental illness is, in effect, culture-dependant and specific (Obeyesekere 1985), we prefer a more nuanced approach. It is by now increasingly well documented that social and cultural factors play a significant role in the presentation and course of mental illness, including those disorders whose etiologies are considered to have a strong biogenetic component (Ware and Kleinman 1992). Ethnography has been recognized by international bodies like the World Health Organization as an important tool in the design and analysis of successful strategies for intervention in the area of mental health (Desjarlais et al. 1995).

What these developments suggest, however, is a reconsideration of any dualism that may have been implied by the analytic distinction between illness and disease (Kleinman 1988; Schweder 1988). Biomedical and psychotherapeutic perspectives are now increasingly subject to the kind of sustained cultural and social analyses that have long been applied to so-called vernacular idioms of illness and healing (e.g., Bosk 1979; Conrad and Gallagher 1993; Martin 1987; Young 1995). The claims made by some practitioners of these disciplines to an acultural and transparent objectivity have been called seriously into question by studies in the anthropology, sociology, and history of science and medicine.

At the same time, current research has seen a renewal of interest in the deep interconnection between human psychophysiology and the social world. Researchers in medical anthropology, for instance, have been moving towards a conception of ‘local biologies,’ by which is meant the flexible interaction of social, cultural, and biological realities to produce locally distinctive patterns of bodily response (Kleinman and Becker, forthcoming; Kleinman and Kleinman 1994; Lock 1993a, 1993b). In a sense, the culture concept has become an even more powerful analytic tool than it was once thought. The realm of biomedical disease, previously bracketed from anthropological concern, is now being relinked with illness in a more subtle and less dualistic fashion.

Medical anthropology’s impressive success with the culture concept has not, however, been matched by an equal concern with the experiential dimension of illness, which is the key to any nondualist conception. There are relatively few ethnographies of experience to match the thick description of culture that has characterized anthropological writing. This is a serious gap, because as sophisticated as our theories of culture may come to be, they will never grant a sufficient theoretical purchase on the complex illness experience of a woman like Mavis Williams, whose life is embedded within, but not determined by, her cultural repertoire. Cultural knowledge alone will never allow us to predict, or even to describe, the outcomes of medical and policy interventions in the lives of real people and communities.

In some cases, a focus on culture may even obscure more than it reveals, allowing the heavy hand of the interpreter to rob lived experience of its forcefulness and contingency by forcing it to conform with preconceived cultural categories. Recourse to these typically timeless categories may involve a denial or depoliticization of fluid social and political experience, which typically includes real struggle over both meaning and material resources. In the case of the Ethiopian-Israeli protesters with which we began this chapter, public health experts, politicians, and anthropologists all tended to reduce immigrant anger to an expression of the symbolic importance of blood ‘in their culture.’ By doing so, they ignored the politically ramified deployment of that symbolism in ways that could not have been predicted through a purely ‘cultural’ reading. At stake for these immigrants was not just the cultural symbolism of blood or of ‘honor,’ but their threatened membership in a national collective, and the right to a voice in the shaping of public health policies whose implementation would certainly have life and death consequences for their community. Culture, in the sense in which it is often invoked in public discourse, is an essential but insufficient component of this discussion.

Even a sensitive and compelling cultural analysis such as that of Anne Fadiman (1997), in her account of tragic miscommunication between a Hmong refugee family in Merced, California, and their American doctors, relies on a totalizing framework of cultural difference, to the extent that other aspects of social life are neglected. A concern with experience, by contrast, would force us to challenge the sovereignty of cultural models, and to dispense with the false sense of predictability that they may help to generate. Despite frequent anthropological critiques, we maintain that the experience of illness is just as irreducible to the theoretical categories of cultural analysis as it is to those of biomedicine (Kleinman and Kleinman 1991). Experience itself must be subject to sustained analysis.

What Do We Mean by Experience?

One of the great challenges facing medical anthropology today is the definition of experience in a useful and consistent way. In this chapter, we are using experience to represent the intersubjective, felt flow of events, bodily processes, and life trajectory which always takes place within a social setting. Experience is ‘the outcome of cultural categories and social structures interacting with psychophysiological processes such that a mediating world is constituted’ (Kleinman and Kleinman 1991: 277). This stands in contrast to commonplace assumptions in everyday usage, which treat ‘experience’ as a property of deep and individual subjectivity. When we say that experience is ‘intersubjective,’ we mean that it is constituted in social space; we call attention to the link between the social world, subjectivity, and psychophysiological processes, which becomes most apprehensible—and for sufferers most intractable—through the experience of illness itself.

Take the case of Huang Zhenyi, a worker from a rural town who was in his late twenties when he was first interviewed. (This account is paraphrased from that published in Kleinman 1986). Huang Zhenyi suffered from depression, and from chronic headaches and dizziness that he attributed to a traumatic childhood experience during the Chinese Cultural Revolution, about which he was able to talk only with his wife. As a child, he had found a piece of paper in his schoolyard with the message ‘Throw down Chairman Mao’ written in bold characters. A friend convinced him to turn the sign in to commune leaders, who responded by calling in public security agents who accused him of the crime. Huang Zhenyi was threatened during a long interrogation that he would not be allowed to go home or to the toilet unless he confessed.

He told his interrogators, however, that he had not written the slogan, and was angry at his friend for not backing up the story. Eventually, he was allowed to return home, and assured his distraught mother that he had done nothing wrong.

This was not, however, the end of Huang Zhenyi’s ordeal. He was arrested again the next morning, and told that this time he would never leave the interrogation room without a confession. Terrified, he signed the confession, but decided not to tell his mother the truth, fearing it would only cause her trouble. Instead, he told her that he had written the slogan himself. ‘If I knew you’d end up like this,’ she said amidst curses and tears, ‘I wouldn’t have wanted you.’ Huang Zhenyi remembers weeping, but found himself unable to tell his mother the truth. ‘I felt like a coward. I couldn’t tell her.’

This experience is not without precedent in Huang Zhenyi’s autobiographical narrative. In fact, it recalls earlier incidents both in terms of the moral meaning that it conveys and the particular nexus of bodily and emotional complaints with which it is identified. At the age of 8, for instance, Huang Zhenyi had played truant from school with some friends. The angry teacher punished them by locking them in a small room, but they escaped. Huang Zhenyi recounts his experience of being searched for by his outraged teacher: ‘I was so frightened, I froze in my place. I could not move.’ When he returned to school the next day, in fact, the teacher ordered him to do menial work around the school rather than study. When he refused, he was criticized severely in front of other teachers. Recounting this event many years later, and in the light of his subsequent experience, Huang Zhenyi reports: ‘My liver became small, and I became frightened, cowardly.’ From this time on, he felt ‘paralyzed’ whenever he had to ‘stand up’ for himself before adults, a bodily symptom of an essentially moral failure.

Because of his admission of guilt in writing the anti-Maoist slogan 12-year-old Huang Zhenyi (again feeling ‘paralyzed’) was marched through the town wearing a dunce’s cap, carrying a sign around his neck on which he had written a self-criticism. He was cursed and spat at by local people, and then sent to do the work of an adult as a peasant on a local production team. In addition, he had to undergo public self-criticism each day, and at one such session he felt himself go numb and mute, unable to scream out the truth he so desperately wanted to tell. Even later, when he returned to school and then resettled in a town where nobody knew his background, he remained unwilling to confess the full story to his mother, and so she died without knowing of his innocence. Huang Zhenyi returns to this fact again and again as a palpable explanation for his later feelings of desperate shame and self-hatred and for his physical symptoms. He felt a searing sense of injustice and anger, which he associated with a burning sensation in his head, dizziness, and exhaustion: all culturally salient symptoms of shenging shuziruo, or neurasthenia. He remained fearful that the events of his youth would return to cause him problems in his current life.

This story cannot be understood without recourse to key cultural idioms of distress and complaint in China, which have already been explored at length elsewhere (Kleinman 1986; Kleinman and Kleinman 1991). It is important to emphasize, however, that cultural analysis alone would be insufficient for the task at hand. Illness experience connects the social and cultural context and the biography of a person—not any person, but a highly specific one—to the disease process. Illness experience is given a shape and pattern by the shared categories of culture, gender, ethnicity, social position, and age cohort, but it is also unique in each particular case. The illness experience is as complex and subtle as the composition of a symphony organized around the simple structure of movements and a basic recurrent theme. Individual perception, interpretation based in past experience, learned patterns of coping, and local explanatory models, as well as the personal virtuosity of metaphor are all brought to bear on the experience of illness. Thus, illness experience becomes a site for the infolding of the social world onto the body—but in a way that interacts complexly with, rather than effacing, the contingency of individual lives.

What is at Stake?

While experience remains a difficult and somewhat elusive subject of inquiry, one of the approaches that has proven most useful has been the use of ethnography to elucidate what is at stake for people and groups in particular local settings. ‘What is at stake?’ is a gloss for attempts to explore the link between culturally patterned and personally contingent elements of experience as they impose themselves on people’s lives. In Huang Zhenyi’s story, part of what was at stake was the culturally interpreted ‘dizziness’ of moral and social disorder that he had been forced to endure. These social conditions continued to express themselves through the embodied symptoms of his illness even many years after the events in question. Huang Zhenyi’s vertigo was (among other things) an expression of stylized meanings carried by that symptom in classical Chinese moral and physiological theory, and cannot be readily understood without reference to them.

At the same time, however, Huang Zhenyi’s story participates in that of an entire generation who lived through the Cultural Revolution. The collective experience of his generation gives Huang Zhenyi’s complaint a resonance that contributes to its lived, phenomenological quality—its reception within a local social network, and its moral meaning for both Huang and others with whom he might share it. Finally, there is a resolutely personal quality to this story that cannot be replicated, and which resonates within a specific life trajectory of loss, disappointment, and the inability to speak at crucial junctures. Failure to attend to any of these levels—the personal, the collective historical, or the broadly cultural—would be to misrepresent the illness experience in its complexity. We will return to this tripártate model at the end of this chapter.

Reflecting on what is at stake for people is one way to avoid the reductionist determinisms to which both medicine and culture theory may sometimes fall prey. In Bali, for example, the typically placid and unemotional demeanor of many who have suffered personal loss led a generation of ethnographers to suggest that grief in its Western sense was foreign to Balinese sensibilities—an imposition of foreign cultural and moral values on an incommensurate cultural reality. Uni Wikan, however, has employed the ethnography of experience to argue that Balinese poise and emotional calm in the face of loss is actually a strategy for avoiding the harm to self and community which are associated with extreme negative emotion (Wikan 1988, 1990). Through her ethnography of what is at stake for particular people in particular life contexts (which is something very different from the ethnography of Balinese ‘culture’ in a generalized and abstract sense), Wikan shows that Balinese are characterized less by a lack of powerful emotion than by a social concern for the devastating power of emotional turmoil. This revision of received wisdom was made possible because Wikan looked beyond the cultural stereotypes in which informants and anthropologists had both sometimes traded, in order to explore what was deeply at stake for a particular young woman in her own long struggle with grief over the death of a lover. The focus on what is at stake in emotional life has special relevance for an ethnography of illness experience, which is often characterized (and in the case of depression, clinically defined) by its powerful affective dimension (cf. Lutz and White 1986; Rosaldo 1989).

Furthermore, the stakes of experience almost inevitably include political and socioeconomic concerns. These may range from considerations of belonging and exclusion for migrants moving through the bureaucratic labyrinths of state medical and immigration services (Ong 1995; Seeman 1997), to the ‘search for respect’ by young men in one of America’s devastated inner cities (Bourgois 1995), or the desire to attain a degree of culturally validated autonomy from crippling chronic pain through employment (Good 1992). The stakes of experience involve whatever is most intractably ‘given’ in a local setting; that which imposes itself upon consciousness and informs social interaction because of its overwhelming importance to the people involved. The stakes are culturally patterned and interpreted, but often of literally life and death significance. As researchers, a concern with the stakes of experience also forces us to consider noncognitive features of social and cultural life: not just how people interpret, categorize, or impose meaning on the world, but how they ‘struggle along’ (Desjarlais 1994), strat-egize, or just make do with the social, cultural, and material resources at hand. The stakes of experience represent a moral engagement with the social world, which is never more stark than in our encounter—as researchers, healers, witnesses or sufferers—with illness.

Ethnography and the Experience of Illness

Explanatory Models

‘Ethnography,’ understood here in an anthropological mode, includes participant observation, interviewing, historical interpretation, use of focus groups, and related qualitative methods, and is a set of key methodological tools available for the engagement with, and analysis, of illness experience. While it may be complemented (and should ideally be informed by) epidemiological, biomedical, and quantitative sociological data (see Inhorn et al. 1990), it is ethnography that primarily lends itself to exploration of the ways in which illness intersects with a particular culturally constructed life-world and a particular life trajectory to produce a unique and irreducible constellation of experience. Of course, there are still various ways in which this tool can be applied.

Some of the early work in medical sociology, associated especially with the work of Talcott Parsons, modelled the illness experience as a ‘sick role.’ The sick role represented the abridgement of social functions and the imposition of social obligations based on a social actor’s recognized illness. The person occupying the sick role would be temporarily freed from the responsibility of normal social duties (in the workplace, or in the family, for instance), but would be expected in return to cooperate with culturally valorized regimes of treatment, and to demonstrate a desire to ‘get better’ in certain culturally specific ways. The unspoken pact between ill person and community is important for its foundational role in negotiating the terms of interpersonal illness experience in a local world. Parson’s student, Reneé Fox, demonstrated in several classical accounts that uncertainty was central to the sick role, its social positioning, and the management of experience in medical care (see Fox 1959).

The perception, conceptualization, and expression of symptoms and coping responses received further development in the work of medical sociologist David Mechanic (1982). Mechanic’s work enabled researchers to operationalize and study particular elements of the experience of sickness, such as how cognitive processes of attending, perceiving, labeling, and interpreting symptoms were influenced by cultural orientations or by the social experience of severe economic constraints. This led, in turn, to research by numerous scholars on the choice-making patterns of those who are ill or seeking treatment, and on patients’ evaluations of their relationships with clinicians and the quality of care they received. These sociological approaches helped to focus attention on the microprocesses of the social construction of illness and healing in particular social and cultural settings, which are an important part of illness experience.

Another means for assessing the influence of cultural and social factors upon a given episode of sickness has been the ‘explanatory model’ approach, which provides a structure for rapid and highly focused ethnographic investigation, and has been of interest to both social science researchers and clinicians for this reason (Kleinman et al. 1978). As the term implies, explanatory models (EMs) concern the ways in which an illness episode is interpreted and understood by patients, healers (including, of course, medical practitioners), and other members of the local social world. It includes notions of etiology, expected course, and predicted outcome, and ideas about appropriate treatment. In addition, EMs may be either shared or contested by differently positioned social actors, and may appear partially incohate or incompletely systematized. Patient EMs may include religious ideas of causality (i.e., karmic retribution or divine judgement), culturally specific understandings of bodily processes (i.e., maladies such as ‘semen-loss,’ ‘tainted blood,’ ‘nervios,’ etc.), and moral committments as to what constitutes an acceptable outcome for treatment, all of which may sometimes be at odds with the EMs brought to bear by physicians, family members, or other interested parties.

It is important, in fact, to resist the conception that EMs are fixed and static, and removed from the continual negotiations that characterize most social life. It is the negotiability of explanatory models that actually helps to impart their clinical usefulness, once identified by a caregiver. The doctor who understands his patient’s explanatory model may be better able to negotiate a shared understanding with that patient as to the requirements and possibilities for treatment (Kleinman 1988). He may also be in a better position to empathize with that patient, and to bridge incommensurate understandings. There are, therefore, several basic questions that will always need to be asked in assessing the explanatory model, which is operative for any social actor at a given time. These questions include those listed below.

  • What is the problem? Is it an illness? If so, what kind of an illness?
  • How does it affect the body-self?
  • What can be expected to happen next?
  • What will be the long-term outcome? Will it get better or worse?
  • What is most to be feared about this condition?
  • What treatment is most appropriate?
  • What is most to be feared about the treatment?

The EM is designed to help caregivers attain relatively rapid access to the experience of illness on the part of patients, and cannot substitute for more comprehensive explorations of the social history of illness and the life histories of the persons involved. It is, however, a start, and in the context of extreme time-space compression (cf. Harvey 1990) that characterizes so much of medical care today, it is an important alternative to the anonymous and increasingly homogenized approach which systematically pushes cultural, moral, and biographical meaning in the illness experience aside. In addition, it is to be hoped that the EM will be understood as two-directional, which is to say that clinicians will learn to apply the analysis of explanatory models to the culture of science and biomedicine, and to the local cultures of their own clinics—and not just to the ostensibly less rational ‘folk beliefs’ and assumptions made by patients.

The self-reflexiveness presumed by this conception of explanatory modelling requires genuine respect for alternative ways of thinking, feeling, and being ill in different local worlds. EMs are inseparable from considerations of what is most at stake for participants in illness encounters, which means that they are inherently moral formulations and should not be ranked with regards to supposed rationality or degree of correspondance to ‘objective’ medical understanding. EMs may or may not square with current scientific understandings and may be infused with moral and explanatory significance drawn from different healing traditions, religious cosmologies, theodicies, and teleologies. In short, the EM should be viewed as a point of entry into the life-world of the sufferer and not mistaken for a way of categorizing people, or explaining away their—cultural’ beliefs. It is not enough to identify patient EMs within a fixed and culturally determined repertoire; the EM is an abstraction for much more dynamic social, personal, and moral processes.

Illness Narratives

Despite the usefulness of explanatory models, they are relatively constrained and artificial when compared with the ‘illness narrative,’ as this concept has been developed by medical anthropologists. Compared with EMs, the illness narrative is a more open-ended approach to the stories that always surround episodes of illness and healing for those who live through them. Huang Zhenyi’s illness narrative, told as an adult suffering from dizziness and headaches, began with stories from his early childhood and adolescence which were, for him, intimately bound up with his current malady. The illness narrative may relate more broadly than the explanatory model to the interpersonal context and effects of illness. It may, in fact, extend to whatever the teller of the narrative considers to be at stake for herself in that episode. Whereas the EM may be constructed on the basis of a relatively narrow set of questions and takes an abstract form, illness narratives require a less structured form of interview technique. The hopes, fears, personal history, and cultural tradition of the ill person are all brought to bear, and it is up to the listener to adopt an analytic stance that is broad enough to accommodate these.

In addition, there are important narrative strategies and poetic conventions that need to be taken into account: voice, genre, master metaphors, and rhetorical strategies are all essential to the way in which illness narratives are constructed. Illness narratives do not simply recount a series of disconnected events, they tell a story that is typically, as in the case of Huang Zhenyi or Mavis Williams, a moral one. Illness is located within the autobiography of the person or community in question. It often contains implicit or explicit teleologies and moral cosmologies that may transcend or lie at odds with rationalized biomedical conventions. If this approach seems to require a developed literary sensitivity as well as personal empathy from the researcher or clinician, that impression may not be inappropriate. The listener needs to be able to hear the story that is really being told and to trace the ways in which widely shared cultural conventions are taken up as the embodied metaphors of some person’s life or illness.

Towards this end, medical sociologist Arthur Frank (1995) has suggested that illness narratives in North America can be divided into four typical genres, the identification of which is essential to determining the meanings that underly a given narrative. As some historians have argued (White 1978), the choice of narrative genre that a writer or storyteller adopts may be broadly determinative of the kind of story he is likely to be able to tell. Frank’s narrative genres are described below.

  • The restitution narrative emphasizes positive responses and outcomes; it is a story of coping with illness, rebuilding the body-self, and remoralization. It may also be evoked in the construction of patients or, even more, of doctors as heroes of the illness experience. These are stories with happy real or projected endings.
  • The chaos narrative, like that of Mavis Williams, tells a story of disorder, distortion, and fragmentation. Anguish, threat, and uncontrollability are the characterisitcs of this type of story. In her ethnography of the victims of ethnic violence in India, anthropologist Veena Das (1994) suggested that this type of narrative will tend to be chosen by those groups in society who actually have lost (or, more correctly, been denied) control over the events that define their lives and deaths. Those in power, argues Das, including localized power over family members and the like, are more likely to invoke systems of meaning in which those who suffer can be made to bear some measure of responsibility for their own suffering, especially when this deflects criticism from structures of social oppression. It is unlikely that this will always be the case, of course; guilty parties may also invoke chaos and meaninglessness to cover the tracks of their concupiscence. Still, it is good to keep Das’s analysis in mind, and to remember that choices of narrative genre may be grounded in the power politics of local settings, and cannot always be taken at face value.
  • Quest narratives emphasize either the search for cure (sometimes expressed through the turn towards experimental treatment or non-biomedical alternatives) or the search for meaning and transcendance within and through illness. The quest narrative is powerfully represented in contemporary North American film and literature, and in many first-person accounts by the ill or by doctors and other healers.
  • Testimony is an explicitly moralized form of story that seeks to bear witness, or to give voice to, those who suffer. Testimony may be organized around the special meaning that a person finds within illness or healing, or the special message to others that their experience is thought to bear. Some narratives of religious healing in the charismatic tradition are especially clear in this regard (Csordas 1994), but testimony may also take the more prosaic form of witnessing to family, friends, and caregivers what it is that has mattered most to a person during his or her illness, or during his or her life. This is a form of narrative that clinicians and caregivers are likely to hear when they elicit illness narratives from those who suffer extreme, or chronic pain, but who may never before have been asked to share their experience of suffering (Kleinman 1988).

These four categories should be understood as guideposts rather than rules to which every narrative must conform. Individuals’ narratives may even participate in different genres at different points in their telling. In addition, it is important to understand that this shifting emplotment of different narrative genres may actually help to tranform illness experience over time (see Garro 1992; Good et al. 1994; Mattingly 1994). From a clinical perspective, it is important to understand what conventions of genre a patient may be presuming and how that may affect her outlook. A tendency to employ quest narrative strategies may, for instance, encourage doctor-shopping or unrealistic demands for experimental and unorthodox treatments.

At the same time, clinicians should realize that their own stories about illness and healing—the stories that they tell themselves, each other, and patients or their families—are also emplotted in crucial ways. The idea that patients should always be hopeful, that physicians should always be heroic, or conversely that ‘realism’ requires a matter-of-fact stoicism on the part of the families of the terminally ill, are all narrative strategies that have bearing on the ways in which care is organized, resources distributed, and meanings either allowed or disallowed from the clinical encounter. Illness narratives are scripted in negotiation between clinicians, patients, and others in a complex interplay that may be compared with literary creation, and which admits of different voices not just for different individuals, but for actors differently placed with regard to gender, ethnicity, or social and political position (Morris 1991).

Each type of illness narrative, furthermore, may hold coded political meanings in a local social field. Illness narrative as testimony may call attention, implicitly or explicitly, to abuses and injustice that exacerbate or are thought to lie at the heart of embodied psychosocial distress (Kleinman and Kleinman 1994). A narrative of restitution, by contrast, may carry messages that include the power of the individual to transcend suffering by following culturally approved rules of action, or the power of the collective (as in nationalist narratives of regeneration) to heal the social body which has been rent asunder. The clinician, like the social scientist, needs to be sensitive to the broad range of meanings at different levels that may be conveyed through illness narratives, or packed tightly within deceptively simple explanatory models of illness, suffering, and healing.

While most clinicians will not be able, given the constraints on their time and resources, to elicit a full illness narrative from each patient, the sensitivity developed through doing so occasionally may help to balance the tremendous pressures placed on caregivers in many settings to spend as little time as possible with each sick person, and to rely on narrowly biomedical categories. Such categories may be highly replicable across life-worlds, but they often, for that very reason, lack the validity that can only be attained through critical knowledge of some particular setting and set of life trajectories. Therefore, even a relatively rapid ‘mini-ethnography’ geared toward the elucidation of explanatory models may be useful in broadening one’s therapeutic approach. As mentioned above, it can lead to enhanced negotiation with patients over models for care that will meet the requirements of treatment without undue violence to the life-world of the sufferer. It can also improve the quality of care by allowing cultural and biographical realities to be taken into account in determining treatment strategies. Perhaps most importantly, this strategy can help us to avoid the dehumaniz-ation—of clinician, researcher, or patient—that comes from treating those who suffer as if they have nothing to say of any relevance to their own lived context.


Representation and analysis of illness experience remains a challenge for theory in medical anthropology. Recently, Kleinman (1997) has suggested a model that can be represented in the form of a triangle: along one side lies cultural representation—the collective patterns of meaning that inform art, theodicy, and other cultural forms. Along the second side of this triangle lies collective experience—the events and social processes that help to define the lives of whole generations of people: World War I, the Cultural Revolution, and the kinds of events that led historian Eric Hobsbawm (1994) to describe the twentieth century as an age of ‘catastrophes and extremes.’ However, collective experience also represents the local events and processes that come to define important elements of the life-worlds of small social networks. The illness narrative of Huang Zhenyi represents these two kinds of collective experience—local and global—in their powerful intertwinement and indeterminacy.

Finally, the third side of this triangle is that of subjectivity, the somato-moral dimension where the expression of illness typically occurs. These three dimensions—cultural representation, collective experience, and subjectivity—exercise a reciprocal influence. Transformation in any one of them affects each of the others, so that observable changes in collective experience during this century, and in collective representations of suffering through the unprecedented commercialization of images of suffering which takes place through mass media (Kleinman and Kleinman 1996), lead us to consider the possible transformations in subjectivity to which these may be linked. The ‘personal experience of illness’ is thus inextricable from the pull of social and cultural change that takes place on every level, from the distinctively local to the increasingly global. It certainly cannot be limited to the health clinic, or to the sickbed of the individual patient alone. It is also a moral experience, and therefore a social one.

The ethnographic focus on experience which we have advocated here suggests these emotional-moral processes that link what is at stake in collective life with what matters in the inner life of patients, and allows us to describe a domain that co-determines the moral content of medical practice together with the institutional and personal constraints of doctoring. This makes illness experience an essential component in any consideration of medical ethics, from which it has too often been excluded or ignored (Kleinman 1995: 41-67). Illness experience, which takes place in the interpersonal spaces of social life, can help to inform ethical deliberation concerning the diversity of possible or actual ways of being in the world and, no less importantly, forces our attention on the powerful constraints-economic, political, and cultural—that stand ready to frustrate the builders of abstract ethical systems among us. Critical analysis of the relationship between human diversity and constraints on human life-worlds should represent an important future contribution by medical anthropology to the consideration of clinical contexts and their ethical, as well as socio-moral, ramifications.

We have argued that illness experience leads simultaneously to an exploration of broad social contexts and embodied states. Ethnography has already done a great deal to document and describe the links between the two, and to suggest theoretical perspectives for understanding that link. Our choice is no longer between narrow culturalist accounts and objectifying medical narratives. Indeed, both of these trends tend to obscure what is really at stake in lived experience and so create a false (in some cases dangerously false) object of analysis. On the other hand, the actual processes through which collective experience, cultural representation, and embodied subjectivity are joined remain relatively opaque; they are easier to trace through careful ethnography than to explain discursively, and this has been frustrating to many. We can only encourage other disciplines, besides medical anthropology, to turn their energies toward the unravelling of this problem. We would suggest though, that any attempt at explanation must follow close along the contours suggested by analysis of complex and personally ramified social experience, wherever it may draw particular disciplinary inspiration. Moral experience cannot be removed from the equation, and least of all from the equation of illness. For clinicians and researchers alike, moral experience is the fulcrum of body, self, and social world where illness matters most.