DEBATE DEBATE

Introduction

In the various disciplines comprising the so-called health field, we observe timid attempts at conceptually constructing the "health" object (Czeresnia, 1999: Levine, 1995), in contrast with the extensive efforts at developing biomedical disease models (Abed, 1993; Berlinguer, 1988; Murphy, 1965; Pérez-Tamayo, 1988; Temkin, 1963), emphasizing the individual and sub-individual levels of analysis.

In order to conduct a preliminary assessment of the conditions allowing for a General Theory of Health, I propose herewith to explore two important underlying dimensions in the scientific field of health: the socio-anthropological dimension and the epistemological dimension. The epidemiological dimension of the health concept was the object of a specific paper (Almeida Filho, in press). Despite recognizing its importance and founding role, the biological dimension will not be covered here, except insofar as it proves indispensable to clarify some specific issue in the health-disease models analyzed herein. Aspects pertaining to the etymology of the term "health" were the object of a related article (Almeida Filho, 2000).

First, I intend to discuss some sociological theories of disease-illness-sickness and biomedical concepts of health, which, despite the limitations discussed below, can doubtless be taken as the point of departure for this undertaking, given the dialectical and multidimensional nature of the health-disease dyad. Second, I will present and assess several underlying socio-anthropological approaches to theories of disease, taking advantage of the opportunity to highlight the semeiologic treatment of health-disease through the theory of "signs, meanings, and health practices". Third, I will analyze several epistemological issues pertaining to the health theme, seeking to justify its status as a scientific object. Finally, I intend to focus the discussion on a proposal to systematize various concepts of health as the initial stage for their application to the theoretical construction of the Collective Health field.

Before entering into the discussion, I should provide a preliminary semantic framework. The English language, the matrix for this specific literature, makes subtle distinctions in meaning between the various concepts of disease and related terms, through two semantic series: disease-disorder-illness-sickness-malady and impairment-disability-handicap. These two series refer to a particular technical glossary, which due to its growing importance in contemporary scientific discourse deserves some attention in the sense of establishing a terminological equivalence in Portuguese, as indispensable background for participation by Brazilian researchers in this debate. Thus, even while recognizing that such attempts can be arbitrary and incomplete, I propose to adopt the following terminology, which I will adhere to strictly throughout the rest of this paper:

• disease = patologia,

• disorder = transtorno,

• illness = enfermidade,

• sickness = doença,

• malady = moléstia.

The irony of social theories of health

In the field of social sciences applied to health, since World War II there has been a somewhat insistent search to objectively define the concept of disease and its correlates (Humber & Almeder, 1997), with a view towards formulating "social theories of health". This section reviews some of these proposals, originating mainly from Anglo-Saxon Medical Sociology.

Talcott Parsons, whose work ascribes an especially central role to health phenomena for an understanding of the social system, defined illness as a "state of disturbance in the normal functioning of the total human individual including the organism as a biological system as much as its personal and social adjustment" (Parsons 1951:431). Parson's theory of the sick role is the first conceptual reference to a series of definitions of the sickness concept as a societal component of the disease-illness complex, as we will see further on. Curiously, Parsons does not highlight the term disease in his theory, rather using illness and disorder, even when it is necessary to refer to the objective pathological aspects of disease (Parsons, 1951, 1964, 1975).

The author later proposed to analyze Health as a social function, defining it as a "state of optimum capacity for the effective performance of (socially) valued tasks" (Parsons, 1964). Parsonian functionalist theory served as the theoretical matrix for approaching individual health as a social role, performance, functioning, activity, and capacity, among others, which were subsequently condensed in the concept of health as social well-being, a characteristic of contemporary "quality of life" rhetoric.

It is difficult to establish who was the first author to systematically postulate a distinction between disease, illness, and sickness. To justify at least a semantic difference between the first two terms, an initial attempt was to rely on common sense. As entries in the traditional Oxford Dictionary (1968), disease means "a condition of the body, or of some part or organ of the body, in which its functions are disturbed or deranged" and illness is simply defined as a "quality or condition of being ill (in various senses)".

Field (1976) conceptualized disease as an abnormality or pathological alteration recognized by means of a set of signs and symptoms defined on the basis of a biomedical conception. On the other hand, illness referred primarily to the subjective experience of an individual's state of "ill health", indicated by feelings of pain, discomfort, and malaise. Paying tribute to the Parsonian theory of the sick role, but without using the term sickness, Field (1976) further contended that illness did not simply imply a "biologically altered state", but also to be in a socially altered state which is seen as both deviant and (normally) undesirable.

In his seminal work Causal Thinking in the Health Sciences, Mervyn Susser (1973) presented two series of definitions that contributed little to overcoming the terminological confusion then prevailing, probably because of the limited diffusion of his writings outside of the epidemiological field. According to Susser, the term disease refers to a pathophysiological process that causes a state of physiological or psychological dysfunction in the individual. On the other hand, illness is an individual, subjective state, a certain psychological and corporal awareness of the disease, while sickness implies a state of social dysfunction in the sick subject, corresponding to Parsons' sick role.

Philosopher Christopher Boorse (1975, 1977) defined disease as an internal state of the body resulting from subnormal functioning of some of its organs or sub-systems. Some such diseases can evolve to illness if they lead to limitations or disabilities that meet the following criteria: (i) that they be undesirable for the subject; (ii) that they be considered eligible for interventions; (iii) that they constitute a justification for normally reproachable social behaviors. Despite the clear functionalist inspiration (along the Durkheim-Parsons lineage), there is no special position in Boorse's original proposal for the term sickness, while illness constitutes a mere subset in the order of diseases, namely those that produce psychological and social consequences for the individual.

Boorse subsequently stated his intent (1977) "to offer a value-free analysis" as the basis for a theoretical concept of Health, along the same lines as the biological concepts of life and death (amongst us Brazilians, the pioneering work of Mário Chaves (1972) had already conceptualized health as an organism's capacity to function within an ecosystem resulting from the Eros-Thanatos opposition, in line of thought intriguingly similar to Boorse's proposal). Boorse proposed a linear articulation involving four basic concepts: "reference class", "normal function", "disease", and "health". The reference class consists of the universe of members of a biological species of the same sex and age bracket. Normal function is defined as an individual contribution that is "statistically typical" in relation to the reference class for the species' survival and reproduction. Disease is a reduction in the "typical efficiency" involved in normal function. Health means simply the absence of disease. Boorse completes his "biostatistical theory of health" with an intentional tautology, indicating that health as a concept can simply imply normality, always "in the sense of the absence of disease conditions".

Contrary to the naturalist theoreticians (mainly Boorse) who believed in an objective and value-free approach to health-disease phenomena, Tristham Engelhardt (1975) identified a fallacy in this operation of considering abstract constructs as concrete things and preferred to treat them as differentiated and autonomous entities. Thus, he justified the definition of disease as a scientific category destined to explain and predict illness, suggesting that the latter, and not disease, was a referent for health phenomena. In his own words (Engelhardt, 1975:137): "Commitment to the concept of disease presupposes that there are phenomena physical and mental which can be correlated with events of pain and suffering, so that their patterns can be explained, their courses predicted, and their outcomes influenced favorably".

Phenomenological approaches to health (Engelhardt, 1975) were critical of naturalist theories' objectivism. Recently, such criticism reached the extreme of challenging the usefulness of the very concept of disease (Hesslow, 1993), apparently with no echo among researchers involved in the important effort at theoretical construction of the field of Medical Sociology.

Seeking an alternative to the expanded use of the disease concept, which spawned confusion of both a logical and semantic order, some authors (Clouser et al., 1997; Culver & Gert, 1982) proposed to adopt the more generic concept of malady. This concept supposedly denoted the universe of categories referring to damage or threats to individual health, including both the various classes of disease, illness, and sickness, as well as the events, states, and processes that were difficult to classify as sickness or disease, such as disorder, dysfunction, dependency, defect, lesion, trauma, etc. Despite the proposal's proper intentions, the concept of malady has not been incorporated into either the theoretical discourse of the Sociology of Health or the technical discourse of Clinical Medicine, and is mostly referred to as a curiosity indicating the insufficiency of the disease concept.

The Pörn-Nordenfeld theory (Pörn, 1984, 1993; Nordenfeld, 1987, 1993), developed as part of an effort at an economic and philosophical justification for Scandinavian "neo-welfarism", was intended to recover a pragmatic definition of health based on updating and correcting Boorse's biostatistical approach. Despite the conceptual limitations and even a certain philosophical naiveté in this formulation, the proposal by Pörn (1984) is certainly interesting, i.e., that the symmetrical opposite of health is neither disease nor sickness, but rather illness (Pörn, 1984). In other words, health is not the objective absence of disease, but the nonexistence of illness in terms of adaptation of a human organism to a biological and social environment (Pörn, 1993).

Presenting a very well-structured theoretical formulation which he entitled the "phenomenology of health", Nordenfeld (1987) proposed a distinction between objective and subjective illness which, as a logical consequence of Pörn's health-illness continuum, leads to the mirror concept of "subjective health". Objective illness is defined by the potential functional capacity not affected by the cause of the disease, while objective health corresponds to the actual exercise of this functional capacity. According to this scheme, subjective illness (or non-health) has two components: (i) the awareness of illness (in the author's words, the "mere belief or awareness that someone is ill") and (ii) the feeling of illness (or the "set of mental states associated with illness"). Thus, as postulated by Nordenfeld, a person P is subjectively healthy if and only if he/she (1) is not subjectively ill, (2) believes that he/she is healthy, or (3) is not experiencing a mental state associated with some currently existing objective illness (Nordenfeld, 1987, 1993).

Along this same line, Fulford (1994) contends that not even the concept of disease is value-free, defending a pragmatic approach through the use of two different levels of analysis, one descriptive and the other interpretative. Given that the former level incorporates disease concepts in which a high degree of consensus prevails, according to Fulford it is necessary to focus more on the latter analytical level. In this case, disease concepts could be referred to generically as "failures". Disease would correspond to a "role failure", while illness would result from an "action failure". Finally, Fulford (1994) challenges the existence of a deterministic link between disease and illness, as postulated by the majority of the authors reviewed, indicating that the actual illness experience cannot be explained by disease concepts, that rather it must be understood as phenomenologically given.

Recently, Boorse (1997) self-critically admitted the need to overcome his negative-evolutive concept of health (based on the dysfunction-disease-illness gradient), proposing to replace it with the notion of "degrees of health". This entails an extremely narrow definition of positive health as the maximum possible degree of health as opposed to any reduction in optimum normal function for the reference class. According to this concept, normality has three levels of specification: theoretically normal, diagnostically normal, and therapeutically normal. The logical opposite of the disease concept would be theoretical (or conceptual) normality. The respective antagonists would fit with the other levels of normality: diagnostically abnormal and therapeutically abnormal. Finally, Boorse analyzes the extreme situations of "illness" (as opposed to "wellness") and death-life. The underlying relationships of belonging and opposition in this interesting scheme are found in Figure 1. Curiously enough, the essential framework of this proposal had also already been laid out in the study by Mário Chaves (1972) quoted above.

A partial inventory is in order at this point. To begin, nearly all of the authors and schools reviewed thus far present proposals marked by a predominantly biological frame of reference. Therefore, they almost inevitably lead to theories not of Health, but of pathological processes and their correlates, in which Health is necessarily seen as the absence of sickness. Consequently, one observes an emphasis on the sub-individual and individual levels, where pathological and experiential processes actually occur. This chain of logical omissions, entailing a reduced focus on the concept of sickness and the sick role impedes a collective conceptualization of Health (except, of course, as the sum of individual absences of disease). Here we find a flagrant irony: despite the promising Parsonian debut, Medical Sociology has not proven capable of constructing a social theory of health.

Anthropological models of disease-illness-sickness

This section expounds on the issue of health-disease models from the interpretative perspective of contemporary Medical Anthropology, converging on a proposal integrating the concepts of disease, illness, and sickness.

Arthur Kleinman, Leon Eisenberg, and Byron Good (Kleinman et al., 1978), seeking to enrich the analysis of non-biological components of health-disease phenomena, systematized in 1978 a model that ascribed special theoretical importance to the notion of "sickness", emphasizing the social and cultural aspects that had paradoxically been overlooked by previous sociological approaches. (Curiously, Kleinman and his disciples omitted prior conceptual developments, even those occurring within the field of social sciences in health as discussed in the previous section). This proposal was based on the distinction between biological and cultural dimensions of sickness, corresponding to two categories: disease and illness. The model is shown schematically in Figure 2, highlighting the implicitly negative definition of health as the absence of sickness.

From this perspective, the pathological functioning of organs or physiological systems occurs regardless of its recognition or perception by the individual or social environment. Within a frame of reference that is quite congruent with Boorse's theory, according to Kleinman (Kleinman, 1980, 1986; Kleinman et al., 1978), disease refers to alterations or dysfunction in biological and/or psychological processes, as defined by the biomedical concept. On the other hand, the illness category incorporates individual experience and perception vis-à-vis both the problems deriving from the disease and social reaction towards illness. The concept of illness thus relates to processes of signifying sickness. In addition to their cultural aspects, meanings also touch on particular symbolic aspects forming the illness itself within the individual psychological sphere, as well as the meanings created by the patient while dealing with the disease process (Massé, 1995).

Subsequently, Kleinman (1988, 1992) partially reviewed his original objectivist position, contending that both disease and illness are social constructs. Illness means the way sick individuals perceive, express, and deal with the process of becoming ill. Illness is thus prior to sickness, which is produced on the basis of a technical reconstruction of professional discourse in the physician-patient encounter, through communication around the culturally shared language of sickness. According to this same author (Kleinman, 1980), health, illness, and care are parts of a cultural system, and as such they should be understood through their mutual relations. To examine them separately distorts our understanding of both their respective characteristics and the way they function in a given context.

Kleinman (1986) further proposed that one of the reasons that different healing processes persist within the same society is because they act on different dimensions of sickness. Thus, one must consider different models capable of conceiving health and illness as resulting from the complex interaction among multiple factors at the biological, psychological, and sociological levels, with a terminology not limited to biomedicine. In order to construct such models, one must turn to new interdisciplinary methods, working simultaneously with ethnographic, clinical, epidemiological, historical, social, political, economic, technological, and psychological data.

Byron Good & Mary-Jo Good (1980, 1982), reinforcing the perspective of intra- and intercultural relativism in illness, postulated that the borders between normal/pathological and health/disease are established by illness experiences in different cultures, through the ways by which they are narrated, and by the rituals employed to reconstruct the world that suffering destroys. From this perspective, sickness (and by extension, health) is not a thing in itself, or even a representation of such a thing, but an object resulting from this interaction, capable of synthesizing multiple meanings.

Good & Good (1980) proposed a "cultural hermeneutic model" to understand Western medical rationality. According to these authors, the interpretation of symptoms as a manifestation of the underlying "biological reality" is characteristic of clinical reasoning, since the latter is based epistemologically on an empiricist theory of language (Good & Good, 1980). According to the biomedical health-disease model, clinical practice is supported by knowledge of causal chains operating at the biological level, following a script for decoding the patient's complaints in order to identify the underlying somatic or psychological pathological process. Thus, the model has a double objective: to establish the disease diagnosis and to propose effective and rational treatment. According to Good & Good (1982), ascribing "symptom meaning" to an altered physiological state proves insufficient as a basis for clinical practice, since psychological, social, and cultural factors influence the experience of sickness, its manifestation, and the expression of symptoms.

One of the central points in this "critical reform" process in medical knowledge consists of the distinction between disease and illness. Agreeing with Kleinman, Good & Good (1982) reaffirm that the disease process correlates with or is caused by biological and/or psychological alterations, while illness is situated in the domain of language and meaning and therefore constitutes a human experience. According to these authors, illness is fundamentally semantic, and the transformation of disease into a human experience and an object of medical attention occurs through a process of attribution of meaning. Thus, not only illness but also disease constitute a cultural construct, in this case based on theory and webs of significance comprising the different medical sub-cultures.

Meaning is not the product of a closed relationship between signifier and the thing (in the sense of an objective reality in the physical universe), but of a network of symbols constructed in the interpretative act, which they refer to as a "semantic network" (Good & Good, 1982). Illness becomes an experience with meaning for each particular individual. Even so, it is important to consider the relationship between individual meanings and the network of meanings inherent to each broader cultural context to which individuals belong. Therein lies the notion of illness as a "network of significance", in the sense of a reality constructed through a process of interpretation/signification, based on the plot of meanings that structures the culture itself and its various sub-cultures. Symptoms, full of at least individual meanings, allow access to biomedicine's web of significance, that is, culturally established disease signs in the form of a "syndrome of meanings" (Good & Good, 1980).

In an attempt to develop an approach to the determination of sickness in societies based on analysis of social relations of production, Allan Young (1980, 1982) presents a critique of sickness models as proposed by Kleinman and Good & Good. On the one hand, he postulates that the Kleinman-Good model only sees the individual as object and arena for significant events regarding illness, failing to report the ways by which social relations form and distribute it. On the other hand, while acknowledging the Kleinman-Good model's advances over the biomedical model, Young contends that the distinction between disease and illness is insufficient to explain the social dimensions of the process of becoming ill.

To overcome these limitations, Young (1980) proposes to replace the Kleinman-Good scheme [sickness = disease + illness] with a triple series of categories (sickness, illness, and disease) with equivalent hierarchical levels, albeit granting greater theoretical relevance to the "sickness" component. It is in this sense that Young ends up postulating an "anthropology of sickness" (Young, 1982). Herein, I propose to designate the Young model as DIS Complex (disease-illness-sickness), as represented in Figure 3.

According to Young (1982), although Kleinman emphasizes the social determinants of the explanatory models and Good highlights the power relations in medical discourses and practices, neither actually undertakes an analysis of these aspects in their work. According to Young, medical practices display an important political and ideological component, based on power relations that justify unequal distribution of illness and treatments, as well as their consequences. Therefore, the elements of the DIS Complex (disease-illness-sickness) complex are not neutral terms, but rather entail a circular process by which biological and behavioral signs are socially signified as symptoms. These symptoms, in turn, are interpreted by way of a semeiology that associates them with certain etiologies and that justify interventions whose results end up legitimating them as diagnostic signs of certain diseases. The author further comments that in pluralistic medical systems, a set of signs can designate different illnesses and therapeutic practices that fail to overlap. Social forces are what determine which individuals suffer certain illnesses, display certain sicknesses, and have access to given treatments. Depending on the sick individual's socioeconomic position, the same disease can imply different illnesses and sicknesses and different healing processes.

According to Young (1980, 1982), the concept of sickness should incorporate the process of ascribing socially acknowledged meanings to signs of deviant behaviors and biological signals, transforming them into socially significant symptoms and events. In his own words, "Sickness is a process for socializing disease and illness" (Young, 1982:270). This process of socializing disease - or better still, of social construction of sickness - occurs in part within and through medical systems, linked to society's broader ideological circuits. Young states that this ideological dimension, through different forms of health knowledge and practice, reproduces specific views of the social order and acts to maintain them. In the final analysis, representations of sickness constitute elements in the mystification of its social origin and social conditions in the production of knowledge. According to Young (1980), the translation of forms of suffering (illness) derived from class relations in medical terms constitutes a neutralization process following the interests of the hegemonic classes. That is, through the medicalization process, the ill condition is reduced to the individual biological level, failing to consider its social, political, and historical dimensions.

Indeed, the focus on sickness supplants the emphasis on the individual or micro-social levels (characteristic of Kleinman's approach, for example). However, although it is an important step forward over its predecessors, Young's DIS Complex opens only one possibility for incorporating the Health issue: once again the mere absence of disease-illness-sickness.

In conclusion, one should value the effort at drafting a general theory of health-disease-care, a badge of the intellectual undertaking of these distinguished heirs to the applied anthropology of the 1970s. Even considering the importance ascribed to patients' beliefs and cultural and personal meanings, as well as the proposal for integrating various components of health care systems and their respective explanatory models, the view of these theoreticians towards the conceptual issue of Health is not sufficiently transdisciplinary to broaden the scope of the medical anthropological approach, restricted to the view of Health as absence of illness. The Kleinman-Good and Young models actually remain constrained to curative practices, focusing on the ill individual's return to normal functioning and healthy life, without even entering into the definition of normality or actually analyzing if the Health concept fits into it.

Semeiologic approaches to health-disease-care

Recently, Good (1994) developed a critical semeiologic perspective for the analysis of health-disease models, reevaluating the semantic network concept, identifying two limitations to it:

The first relates to the redefinition of the DIS Complex in light of linguistic theory, given the insufficiency of the perspective according to which a symbol condenses multiple meanings. According to Good, one must recognize the diversity of national, ethnic, religious, and professional languages in the contemporary world, as well as the multiplicity of voices, the individuality of these voices, in short, an inter-dialogue and an alter-dialogue present in the construction of discourses on health-disease. Illness is not only constituted by the individual point of view, but by multiple and frequently conflicting pathways; in this sense it is dialogic. Even while illness is synthesized in familiar narratives, loaded with gender and kinship policies, it is also (and now as disease) objectified as a specific form of physiological disorder in case presentations and conversations among physicians, even if these objectification can be subverted or resisted by patients. Sickness is immersed in a social web in which everyone negotiates the constitution of the medical object and the guidance of the material body.

The second limitation to the analysis of semantic networks refers to the reduced possibility of representing the diversity of forms of authority and resistance associated with the medical system's central elements. Semantic networks, albeit produced by power and authority structures, can provide the necessary means to understand how hegemonic forms are organized and reproduced, since they are culturally rooted and sustain discourses and practices. However, Good (1994) acknowledges that this relationship between semantic structures and hegemonic power relations has not been sufficiently developed by the main authors in this theoretical field, as maintained by Young's radical critique.

The notion of semantic network should thus be expanded to indicate that the meaning of sickness is not univocal, but the product of interconnections. It is no longer just a syndrome of meanings, but also a syndrome of experiences, words, feelings, and actions involving different members of society. This set of elements is condensed in the essential symbols of the medical lexicon, implying that such diversity can be culturally synthesized and objectified. Semantic networks constitute deep structures that link illness to a culture's fundamental values, meanwhile remaining outside of the explicit cultural knowledge and awareness of the society's members, presenting themselves as natural. This new analytical agenda for semantic networks (Good, 1994) treats the DIS Complex as a narrative, both natural and cultural, resulting from concrete, partially indeterminate sickness processes, a veritable script marked by a plot with different perspectives.

Advancing such critical perspective, Gilles Bibeau and Ellen Corin state that contemporary cultural anthropology, through its interpretative and phenomenological watersheds, has proven incapable of dealing with the complexity of health and sickness processes. This necessarily results from the emphasis on the study of subjective experiences in falling ill and the reification of sickness narratives, taken as autonomous texts, without ever establishing relations with either the overall sociocultural context or the disease's "objective" dimension. Despite emphasizing the importance of cultural values and the influence of the semantic network concept in their work, Bibeau, Corin, and collaborators (Almeida Filho et al., unpublished manuscript; Bibeau, 1988, 1994; Bibeau & Corin, 1994, 1995; Corin, 1995; Corin et al., 1993; Corin & Lauzon, 1992; INECOM, 1993) reaffirm the need for a macro-social and historical approach to understand local contexts. This means establishing an epistemological, theoretical, and methodological connection between different dimensions of reality, articulating a meta-synthetic theory or "global perspective" (Bibeau, 1988) intended to integrate essential semeiologic, interpretative, and pragmatic elements for a cultural model of health-disease-care. In the particular sphere of health, the issue is to explore the relations between semeiologic systems of meanings and external conditions for production (the economic-political context and its historical determination) and the experience of falling ill (Corin, 1995).

Seeking to analyze the issue of different levels of determination in health phenomena, these authors (Bibeau, 1994; Bibeau & Corin, 1994; Corin, 1995) propose an analytical scheme based on two central categories: collective structuring conditions and organizing experiences. Using these concepts, they intend to represent the different contextual (social and cultural) elements that link to form the systems of social responses towards "structural pathogenic devices". The structuring conditions encompass the macro-context, that is, environmental constraints, political power networks, economic development parameters, historical legacies, and daily conditions of existence (or modes of life). In other words, it is a matter of conditioning factors acting to modulate culture and limit functional freedom of action at the species and individual levels. Collective organizing experiences, in turn, represent the elements in the group's socio-symbolic universe that act to maintain the group's identity, value systems, and social organization (Bibeau, 1988). Thus, by postulating that semeiologic systems and modes of production link to produce the experience of falling ill, the authors retrieve Young's intent to consider the socioeconomic, political, and historical context in health-disease-care processes.

From this perspective, Bibeau and Corin effectively point to an ouverture of meaning in the health field, implying a new view of the DIS Complex. They propose understanding the falling-ill process as based on the above-mentioned "global perspective", linking individual trajectories, cultural codes, the macro-social context, and historical determination. To this end, they advance an anthropological, semeiologic, and phenomenological theoretical framework to study local systems of signification and action vis-à-vis health problems. Such systems are rooted in the group's social dynamics and central cultural values underlying the individual construction of the falling-ill experience and collective construction of the social production of sickness (Bibeau, 1994; Bibeau & Corin, 1994, 1995; Corin, 1995).

In the communities' spheres of symbolic production, corporal, linguistic, and behavioral signs are transformed into symptoms of a given illness, acquiring specific causal meanings and generating given social reactions, shaping what Bibeau & Corin (Bibeau & Corin, 1994) propose to call the "system of signs, meanings, and practices of health" (SmpH). In general, locally constructed popular knowledge is plural, fragmented, and even contradictory. Popular semeiology and cultural models of interpretation do not exist as an explicit body of knowledge, but are formed by a varied set of imaginary and symbolic elements, ritualized as rational. According to these authors, popular knowledge about Health and its counterparts (expressed in the DIS Complex) are linked and expressed in terms of socially and historically constructed SmpH systems.

SmpH systems thus shape a popular semeiology of health problems in context. To approach them systematically or "scientifically", the authors propose to look beyond the professional diagnostic criteria of the biomedical model and document the particular cases comprising actual cultural variations (Almeida Filho et al., unpublished manuscript; Bibeau & Corin, 1994, 1995; Corin, 1995). In the daily process of defining categories and recognizing cases in these categories, "ordinary" people (the community, according to Bibeau & Corin) do not necessarily function by identifying clear-cut categories of thought, but by perceiving similarities and analogies and establishing a continuity among cases according to a rich and fluctuating range of criteria (Almeida Filho et al., unpublished manuscript; Bibeau & Corin, 1994, 1995; Corin, 1995). Component categories of SmpH systems are fragmented, contradictory, partially shared, locally constructed, organized in multiple semantic and praxeological systems (i.e., structured in practices), in historical context, and accessible only through concrete situations - events, behaviors, and narratives. This mode of categorization refers to object-models formed by "Lakoff prototypes" instead of hierarchical classifications of discrete, mutually exclusive, and stable categories, typified by formally consistent logic. The concept of "prototype", key to linguist George Lakoff's theory (Lakoff, 1993), implies categories of fluid, imprecise meanings with relative degrees of stability, discriminated by fuzzy limits in definition. Because they differ from the categorical logic prevailing in Western, Aristotelian thinking, Lakoff prototypes can be better understood through alternative systems like Zadeh's fuzzy logic (as suggested by Lakoff himself) or Newton da Costa's para-consistent logic (Costa, 1989).

The theory is still being constructed and is thus quite incomplete, full of gaps and inconsistencies. Contrary to the approaches discussed earlier, the SmpH theory unhesitatingly presents itself as the basis for a General Theory of Health. Nevertheless, even in an indirect and attenuated way, this theory is still centered on illness, justified as such by the observation that popular semeiology is also structured on the concept of sickness and its correlates. On the other hand, by considering the biological field underlying the DIS Complex only in a partial and fragmented way, the SmpH approach runs the risk of structuring itself abstractly as a kind of anti-naturalism, prioritizing social, cultural, and linguistic aspects of sickness over the material and objective elements of disease, captured by modern medical technology. Of course, taking medical knowledge and clinical practice as cultural constructs (which they actually are), and consequently as objects of anthropological inquiry, does not shift the material basis of health-disease-care processes and phenomena. This theory merely outlines a broad definition of "structural pathogenic devices" by developing an analysis of different operational levels in the SmpH systems restricted to local-global and social micro-macro polarities, characteristic of contemporary anthropological debate. Any heuristically efficient treatment of the Health issue will certainly have to anchor it in more complex explanatory models and broader conceptual spectra: that of the molecular-subindividual-systemic-ecological in the biological dimension and of the individual-group-societal-cultural in the historical dimension.

The epistemology of Health

What has been discussed thus far appears to shape a certain chronicle of a concept's resistance. "Health" is certainly not a docile or submissive object of analysis. It has resisted more or less competent attempts at domestication by the sciences of both structure and interpretation. A critical inventory of this effort leads one to conclude that the social and anthropological scientific approach to the Health issue has reached its limits, proving incapable of dealing with the properties of the object-model it intends to construct. But is the quest for a General Theory of Health really feasible, taking the health concept as a given object-model? In short, can health be treated as a scientific concept? Or, does this undertaking entail an underlying philosophical problem or some essential epistemological obstacle? If it is possible to conceive of Health as a concept, how can epistemology contribute to the effort? This section is intended to evaluate this set of questions.

Without a doubt, the nature of Health constitutes a secular philosophical question, perhaps of the magnitude of Russell's paradox or Hume's problems. Descartes identified it and Kant later systematized it as a basic problem for philosophy (Canguilhem, 1990). Therefore let us call it Kant's Problem.

Among the contemporary philosophers who have focused on the Health issue, Georges Canguilhem deserves special attention. In his inaugural work Le Normal et le Patologique (1978), Canguilhem indicated that the medical definition of normality stems largely from physiology, founding a positivity that impedes viewing sickness as a new form of life. Therefore, disease could not be admitted as an objective datum, given that positivist scientific methods only have the ability to define varieties or differences, without any positive or negative vital value.

From this perspective, the normality-pathology and health-disease conceptual dyads are not symmetrical or equivalent, to the extent that normal and pathological do not constitute contrary or contradictory concepts. Pathological does not mean the absence of norms, but the presence of other vitally inferior norms, which prevent the individual from experiencing the same mode of life allowed to healthy individuals. Hence, for Canguilhem, pathological corresponds directly to the concept of sick, implying the vital opposite of healthy. Possibilities in the state of health are superior to normal capacities: health constitutes a certain capacity to overcome the crises determined by the forces of disease to install a new physiological order.

Representing a historiographic watershed in the Canguilhemian theory of the normal-pathological tension, Michel Foucault (1963, 1976) sought to indicate how new standards of normality emerged in the sphere of general and psychiatric medicine. In the context of 18th-century cultural reconstruction, attempts were made to intervene in human individuals, their bodies, their minds, and not only in the physical environment, to thereby normalize it for production. To list the normal possibilities for human yields and capacities, as well as the parameters for normal social functioning, became the task of psychiatric medicine, psychology, and applied social sciences. From this perspective, the implicit concepts in Foucault's work reveal his adherence to a definition of health as an adaptive capacity (or submission) to disciplinary powers.

Subsequently, Canguilhem (1966) stated that normality as a life norm constitutes a broader category, encompassing healthy and pathological as distinct sub-categories. In this sense, both health and sickness are normal, to the extent that both imply a certain life norm, where health is a superior life norm and sickness is an inferior one. Health is no longer limited to the perspective of adaptation, no longer unrestricted obedience to the established model. It is more than this, to the extent that it can constitute itself precisely by non-obedience and transformation. According to the elder Canguilhem (1966, 1990), health as the perfect absence of sickness is situated in the field of disease. The threshold between health and sickness is singular, although influenced by forces that transcend the strictly individual, like the cultural, socioeconomic, and political grounds (Caponi, 1997). In the final analysis, the influence of these contexts occurs at the individual level. Nevertheless, this influence does not directly determine the result (health or sickness) of this interaction, to the extent that its effects are subordinated to normative processes of symbolization.

Canguilhem systematizes his reflections on health in a little-known lecture given at the University of Strasbourg in 1988 and published in a limited edition (Canguilhem, 1990). In this paper, after a brief etymological analysis, referring back to Hippocratic ideas, Canguilhem notes that over the course of history, health was treated as if it could not be grasped by reason and thus did not belong to the scientific field. He dwells particularly on the philosopher Kant, who, as we have seen, provided the basis to position health as an object outside of the field of knowledge, whereby it could never be a scientific concept, but rather a commonplace, popular notion, within everyone's reach.

The idea that health is something individual, private, unique, and subjective has recently been defended by the eminent philosopher Hans-Georg Gadamer, one of the main exponents of contemporary hermeneutics (Gadamer, 1996). According to Gadamer, the mystery of health lies in its elliptic, enigmatic character. Health does not present itself to individuals. It cannot be measured, because it entails an internal agreement and cannot be controlled by external forces. Gadamer goes so far as to say that the mystery of health is equivalent to the mystery of life. In his opinion, the distinction between health and illness cannot be clearly defined. The distinction is pragmatic, and can only be accessed by the person who feels ill and who, no longer capable of dealing with the demands of life and the fear of death, decides to visit the doctor. Gadamer's conclusion (Gadamer, 1996) is simple: due to its private, personal, radically subjective nature, health can never be reduced to an object of science.

Canguilhem (1990) would agree that health is a philosophical issue to the extent that it escapes the reach of instruments, protocols, and scientific equipment, since it is defined as free and unconditional. This "philosophical health" would cover, but not be confused with, individual, private, and subjective health. It is a phenomenon without a concept, emerging from the praxeological relationship in the physician-patient encounter, validated exclusively by the sick subject and his/her physician. Clinical knowledge is attributed to the mission of applying a technology and practice of protecting this subjective, individual health. Yet philosophical health does not only incorporate individual health, but also its complement, recognizable as a public health (i.e., a health made public).

The philosopher's notion of public health, referring to ethical and metaphysical questions (which would result for example in the notions of utility, quality of life, and happiness), moves away from the public health expert's concept of health, which understands the state of health of populations and its determinants, both in the sense of a complement to the epidemiological concept of risk and as a reference to the broader concept of the radical need for health. The concept of radical need comes from the post-Marxist Hungarian philosopher Agnes Heller (1986), providing an especially interesting conceptual opening for a General Theory of Health endeavor, to the extent that it implies health as something positive, albeit in the partial sense of filling an essential lack or need in a subject (like resistance or resilience) or society (as a positive health situation) (Paim, 1996). This proposal was applied to the health field by Ricardo Bruno Gonçalves, according to whom "health needs could be conceptualized as what must be achieved for a being to continue to be a being" (Gonçalves, 1992:19) - I owe this observation to Jairnilson Paim (personal communication).

Canguilhem (1990) is against the exclusion of health as an object of the scientific field, anticipating a stance contrary to that of Gadamer. He contends that health is realized in the genotype, in the subject's life history, and in the individual's relationship to the environment; hence, the idea of a philosophical health would not preclude taking health as a scientific object. While philosophical health would encompass individual health, scientific health would be public health, that is, a healthiness constituted in opposition to the idea of morbidity. Since the body is the product of complex processes of exchange with the environment, to the extent that these processes can contribute to determine the phenotype, health would correspond to an implied order both in the biological sphere of life and the mode of life (Canguilhem, 1990). As a product/effect of a given mode of life, health implies a feeling of being able to confront the force of illness, thus functioning as a sort of openness towards social risks, as analyzed by Caponi (1997).

At this stage of his argument, Canguilhem refers to Hygiene, which begins as a traditional medical discipline, made of norms, not disguising its political ambition of regulating the lives of individuals. Beginning with Hygiene, health becomes an object of calculation and begins to lose its dimension as a private truth, receiving an empirical meaning as a set and effect of objective processes. Canguilhem (1990) insists that health is not only life in the silence of the organs, as affirmed by Leriche, but also life in the silence of social relations. It is from this perspective that we can insert the discourse of collective health as we know it. Yet Canguilhem (1990) contends after all that scientific health could also assimilate some aspects of individual, subjective, philosophical health, so that not only sickness and healthiness (or, using a more up-to-date terminology, the risks) should be studied by science. Figure 4 is an attempt at schematically depicting the Caguilhemian position on this issue.

It is curious to note that Canguilhem had already taken a stance on this question long before. On the one hand, he recognizes the health concept's potentially scientific quality, since even admitting that this does not refer to an existence, rather to a norm with function and value, "this does not mean that health is an empty concept" (Canguilhem, 1978:54). On the other hand, young Canguilhem finds no justification for a specific health science endeavor, at least at the individual level. In his own words: "If health is life in the silence of the organs, there is no science of health per se. Health is organic innocence. And it must be lost, like all innocence, in order for knowledge to become possible" (Canguilhem, 1978:76).

In this same sense, the radically phenomenological Gadamerian perspective in defense of private, subjective, inherently enigmatic health would justify ruling out the feasibility of a scientific approach to health. However, I see as a paradox the fact that one of Gadamer's main proposals (Gadamer, 1996) turns out to be crucial for the advancement of an alternative formulation for the scientific object of health. Based on an etymological argument, as is his style, he defends the idea that health is inescapably all-encompassing, because its concept directly indicates wholeness or totality. From this angle, the Gadamerian notion of the "health enigma" ends up opening the way for a synthetic (or meta-synthetic, as we shall see later) approach to the scientific concept of health.

The Argentine epistemologist Juan Samaja, author of the classic Epistemología y Metodología (Samaja, 1994), a rare case of a philosopher with training and interest in Public Health, takes Canguilhem as his point of departure to investigate the conditions allowing for a scientific theory of health. Samaja (1997) criticizes both the Canguilhemian premise that the health concept is concerned fundamentally with the biological world and the implicit Foucaultian premise that proposes a purely social or merely discursive (ideological-political) concept of health.

According to Samaja (1997), the paradigm of Complex Adaptive Systems could serve as the epistemological basis for overcoming the biological-social antinomy, given conceptual demands already established by the development and practical use of the "health" notion in modern lay and technical discourses. In his opinion, one must conceive of the health concept as an object with distinct hierarchical facets, which "allows one to dialectically approach the health-disease dyad and the practices comprising it, leaving room for the recognition of various planes of emergence, in a complex system of adaptive processes" (Samaja, 1997:272).

Incorporating elements from contemporary critical hermeneutics, Samaja proposes that the object-model "health" should operate under four essential ontological determinations:

Normativeness. The health object is normative because it exists in and consists of the hierarchical interfaces in dynamic social and biological systems, both real and ideal, which shape the human world by means of processes involving the establishment and evaluation of norms for existence.

Dramaticism. The health object is dramatic in two senses: first, in the recursive sense, to the extent that it exists in and consists of iterative, reproductive, and transformative processes of the hierarchical interfaces; second, dramatic in a conflictive sense, given that each hierarchical order maintains a high level of autonomy and therefore of vulnerability vis-à-vis the interfaces.

Reflexivity. The health object is reflexive because it exists in and consists of the field of the professed senses and practices experienced through "productive-appropriative (specifically human) conduct".

Historicity. The health object has an onto-socio-genetic nature: it exists in and consists of the dialectic of structural processes that recapitulate past geneses.

In this pathway of construction, which purposely takes health as a social value (and almost as an ideal type), Samaja highlights its complex, plural nature, fundamentally one of linking multiple determinations:

The object of Health Sciences, as a complex object that contains sub-objects with different levels of integration (cells, tissues, organisms; persons; families; neighborhoods; organizations; cities; nations...), entails a large number of hierarchical interfaces and an enormous amount of information, in which its experiences and postulations (both true and false) on normal/pathological, healthy/ill, and curative/preventive acquire meanings and dramatic dimension (Samaja, 1997).

The author derives from these reflections a series of epistemological conclusions, amongst which he highlights that the theoretical health field emerges from the production and formulation of what he calls a "politomorphous" knowledge on the normal-pathological dialectic. To this end, the interdisciplinary field of the health sciences is structured on the cognitive production of the various subordinate objects, revealing different planes of emergence and hierarchical interfaces. The fundamental question in this epistemological investigation consists precisely of the identification of the structuring interfaces in the multifaceted totality of the object-model "health". According to Samaja (1997), the main interfaces of Health are: "molecule//cell (specific category: autopoiesis); cell//organism (category: ontogenesis); organism//society (category: structural coupling)". (Samaja does not refer to an important intermediate interface, albeit one that occurs at a subindividual level, involving organs and systems in the organism, and whose specific category might be differentiation. I owe this observation to Lígia Vieira da Silva, by way of personal communication.) In addition, he proposes to consider the interfaces in the societal sphere, playing out as follows: biosociety//gentilic society; gentilic society//political society. In short, Samaja's contribution is a critical proposition vis-à-vis Canguilhem's thinking, yet one that intermediates it, allowing for its instrumentalization as a frame of reference for a General Theory of Health.

Turning to both the sciences of symbolic systems and those of organized biological systems, Samaja proposes a perspective which doubtless overcomes Kant's Problem and updates Canguilhem's theory concerning the new paradigmatic developments in contemporary science. After all, in Kant's time only physics, astronomy, and natural history were considered science. It is not surprising that for the founding philosopher of modern epistemology, it appeared inadmissible to consider a radically subjective question (like Health, at least at the individual level) as a potential problem for science. We should not forget that scientific psychology and anthropology had still not been created, that social relations, the unconscious, and the symbolic contents of culture and history were still not scientific objects, and that the ethnographic method had not been developed. In addition, clinical practice today is not what it used to be (as for example in young Canguilhem's time). Biotechnology has invaded/trespassed molecules, tissues, organs, the human body. Diagnostic classifications, the genetic code, and the immune system have increasingly become the object of so-called information sciences. A word of caution is thus in order vis-à-vis the radical constructionism permeating any Canguilhem-like analysis, which by appearing to ignore the naturalness of disease, becomes a source of abstract reflection which fails to instrumentalize a consistent critique of the hegemonic medical model.

Modeling health

As discussed above, the biomedical conceptions of health and the sociocultural theories of health-disease present major limitations that reduce their value as a conceptual reference to deal with the multidimensionality of the DIS Complex (disease-illness-sickness). Functionalist medical sociology developed processual models for the social determination of illness that only tangentially allow one to infer health as the result of a daily process of constructing social responses. Neither has medical anthropology ever proposed to define a theoretical category called "health", focusing on the ethnographic specificities of the notion of sickness and its correlates. Despite their theoretical and methodological advances, both perspectives focus on curative practices, and insofar as necessary define health as the absence of illness or sickness.

The thinking originating from Canguilhem's work effectively constituted an epistemological foundation of the utmost importance for developing new theories of health in the field of Collective Health (Caponi, 1997). Nonetheless, the philosophical approaches to the concepts of normality and health, by emphasizing the individual and subindividual levels of analysis, ended up reducing the scope of their contributions.

Despite such limitations and criticisms, all this effort represents an inestimable contribution to theoretical advancement in the health field. In the current essay, it was possible to briefly consider the accumulated heuristic potential in the interfaces between the social sciences and the health sciences, thus the identification of some objective conditions for the formulation of a proposal to systematize the conceptual problem of Health.

From the preliminary exploration of the epistemological foundations of the health issue, one can retrieve the following potentially useful elements for the current proposal:

a) In accordance with anthropology's multivocal approach in Bibeu and Corin's systemic interpretation of health, the plurality of discourses structured with a scientific basis should be contemplated in this process, shaping descriptors capable of ordering the possibilities for the concept's empirical reference.

b) Converging with Canguilhem's stance, selected forms of the "health" concept can legitimately constitute an ontology of health as a scientific object.

c) Respecting the impasse raised by Gadamer, yet retrieving his argument regarding the holistic nature of health, the object-model "health" should incorporate a metasynthetic component into its construction, respecting its integrity-totality.

d) Considering Samaja's contribution, a constructive approach to the scientific quality of "health" should contemplate the field's hierarchical interfaces, organizing the concept's explanatory structures as a heuristic object-model.

Based on the investigation into the health concept in different contemporary discourses, I identify the following background issues indicating theoretical problems that must be overcome:

How to conceptualize health through the planes of emergence of phenomena and processes that define it concretely? Is it possible to define health as a single cross-section, by means of a theory capable of transmigrating from the individual-singular to the collective-social levels?

How to absorb the intuitive notion of health as absence of sickness into a positive concept of health? And how to link this incorporation into the various planes of emergence of health-disease?

How to move towards a positive concept of health, considering the concept's historicity and its applicability as an underlying notion in processes of transformation of a given health situation?

Considering the definitions of hierarchical interfaces and planes of emergence and integrating the contributions by applied social sciences, as reviewed above, I propose an effort at semantic and theoretical specification of what could be called Modes of Health, as shown in Table I. This organizes the terminology used for categories of non-health available to the various health sciences, in addition to distinguishing between the variations in the definitions of normality and health and their potential empirical descriptors.

Like any schematic representation, this one is an attempt at depiction which is necessarily partial and impoverished in comparison to the rich and complex underlying reality. The various modalities of health and the corresponding categories of non-health are organized according to hierarchical planes of emergence: subindividual (systemic//tissual//cellular// molecular), individual (clinical//private), collective (epidemiological//populational//social). What I propose here is a glossary of categories for non-health which in a sense incorporates and expands the preliminary semantic demarcation of disease - illness - sickness. Note that the category "disorder" (or transtorno in Portuguese) occupies a level equivalent to the definition of disease in the clinical sphere.

As in any scheme, I seek to indicate equivalent descriptors for the respective level and sphere. Thus, at the subindividual level, normality and pathology (in the original Canguilhemian sense) correspond to the descriptor "state". At the individual level, in the clinical sphere, normal health corresponds to disease (structural) and disorder (functional), having "signs & symptoms" as descriptors.

At the subindividual and individual planes of emergence, at any level of complexity, the health object can be examined based on an explanatory approach with a determinant basis, producing highly structured causal metaphors. In this case the issue is to produce (or polish) some partial facets of the object-model Health: the biomolecular process in the normal systems or the sustained physiological process in healthy subjects as equivalent to the pathological processes as manifested in the "case", or the "case of illness". The constitution of the Clinical Medicine disciplinary field around this facet of the total health-disease object has been treated both in historical/epistemological as well as praxeological terms (Almeida Filho, 1997; Clavreul, 1978).

Private health, with Gadamerian phenomenology, and individual health, the object of an "epidemiology of mode of life", both refer to the "illness" category, according to the distinction proposed under the Susser-Kleinman-Young line of thought. Note that in each of these cases the descriptors display a certain sense of antagonism: "health status" as the intent to objectify the individual mode of health and "health feeling" as the intimate, particular, private mode of health, which cannot be made public.

Within this scheme, it is also possible to situate the conventional epidemiological perspective (the epidemiology of risk factors), founded on an inductive logic with a probabilistic basis (Almeida Filho, 1997; Ayres, 1997). From this perspective, the health-disease object is reproduced as a specific concept, with risk production models based on the direct action or interaction of risk factors. In the epidemiological sphere of risk analyses, measurement-type quantitative descriptors (rates, coefficients) can deal with the subset's counter-domain [the sick population groups], equivalent to the population residue (1 - risk).

The notion of public health in the elder Canguilhem, which one can call "healthiness" - in contrast with the idea of morbidity in traditional public health discourse, can have "health situation" as an efficient descriptor. Finally the modes of "social health", equivalent to the concept of sickness in interpretative medical anthropology, could be approached through Bibeau-Corin's systems of signs, meanings, and practices of Health (SmpH). Indeed, the SmpH theory provides the possibility of incorporating sickness into the health concept itself, to the extent that it sees the experience of sickness as a way of structuring the social representation of health by constructing subjectivity and the subject's relationship to the material and symbolic world.

A synthesis of this initial approach to the problem of theoretically defining Health is that one cannot speak of health in the singular, rather of various "healths", depending on the levels of complexity and planes of emergence at stake. Such an early conclusion is in line with Czeresnia's (1999) emphasis on the important difference between prevention and promotion regarding the pragmatic use of concepts of Health. However, this issue is best depicted in Sol Levine's insight concerning levels of "health reality", as in the following quote: "But what is health? It is, of course, not directly observable, but is inferred. Health is, first of all, a conceptual construct that we develop to encompass a range of different classes of phenomena [... in] three levels of reality: the physiological, the perceptual, and the behavioral" (Sol Levine, 1995:8).

At this still preliminary stage of exploration and theoretical formulation, there is no doubt that one must face a new family of objects, i.e., object-models not defined by their components, functional principles, and dimensions, which do not prove amenable to the production of knowledge by way of fragmentation (hence, objects adverse to analytical processes). Such synthetic models tend towards a new degree of formal ascension to become meta-synthetic objects, constructed for (and by) reference to the facts produced by the so-called Health Sciences.

Further comments

Before concluding, two questions are in order by way of an overall justification for this essay: Why not adopt the health-as-absence-of-disease perspective, as almost everyone has done? Why seek to construct a positive health concept? Why propose a General Theory of Health rather than a perhaps more realistic unified theory?

The first question has important practical and theoretical consequences. Let us first look at its practical side. Intuitively, it is not easy to propose interventions in a void, aimed at transforming situations that determine absences, powers, or virtualities. To consolidate subjects' resistance and resilience towards the DIS Complex, to induce an increase in what has been termed social capital (Kawachi, 1999; Kawachi et al., 1999), to reinfore the human ties that produce quality of life in daily life through social support networks (Kaplan et al., 1977), in short, to effectively achieve the much-lauded health promotion, we need a specific construct to designate Health (Noack, 1987). This means constructing a positive object-model for knowledge and intervention rather than a negative object, a mere conceptual residue from explanatory modes of biological and social life based on their logical opposite.

The health-as-absence-of-disease perspective, albeit conceptually comfortable and methodologically feasible, cannot fully deal with the processes and phenomena referring to life, health, sickness, suffering, and death at any of the levels of reality identified above by Levine (1995). Just as the whole is always greater than the sum of its parts, health is much more than the absence or inverse of sickness. It is a crucially interesting logical problem, to be solved by overcoming the antinomy between health and sickness inherited from the traditional biomedical model.

Let us return for a moment to Talcott Parsons. In his last work (Parsons, 1978), a detailed analysis of the relationship between social practice and the human condition (and a little-known and poorly evaluated work, even among social theoreticians), this author resumes the theme of Health, defining it "as a symbolic circulating medium regulating human action and other life processes", in the context of a curious analogy with the economic concept of wealth (health = wealth). Like currency, Health does not constitute a value per se, but does in fact become a value in exchange processes. Thus, according to Parsons, Health is not a capacity that is found in the body, nor even does it refer to the individual organism, rather it is a mediator in the interaction between social subjects. Health is not something that can be "stored"; it only exists while it circulates, when it is "enjoyed". Health, as stated succinctly by Parsons (1978:69), "is the teleonomic capacity of an individual living system... the capacity to cope with disturbances... that come either from the internal operations of the living system itself or from interaction with one (or) more of its environments". Health is thus not the inverse or absence of sickness; and sickness (always illness, according to Parsons) should be the "obverse" of health.

The second question haunts other contemporary scientific fields. The basic difference between a unified theory and a general theory is that the former is postulated as a global form of exclusive and all-encompassing explanation, valid for all levels and contexts, while a general theory implies alternative modes of understanding, respecting the complexity of the objects and the plurality of different scientific approaches to an interdisciplinary problem.

The epistemological critique expounded in this paper was highly useful for establishing the central problem of levels of complexity and planes of emergence, indicating that health-disease phenomena cannot be defined as essentially an individual-clinical or subindividual-biological issue. In addition, the objects of Health are polysemous, plural, multifaceted, transdisciplinary, simultaneously ontological and heuristic models capable of traversing (and being traversed by) spheres and domains referring to different levels of complexity.

I wish to conclude by leaving a hint of doubt: and what if Gadamer is right? Perhaps health is more a question of life than of science; if so, then it would not make sense to construct it as the object of even relatively objective knowledge. It may be that metasynthetic and sensitive objects like Health and the DIS Complex can only be found beyond proud science and vain philosophy.

Despite this hint of doubt, it is up to us to proceed. Based on this preliminary essay, we can test hermeneutic methodologies for investigating scientific discourses, assuming them no longer as an external object of inquiry, but seeking to retrace the steps of the thematic investigation of health itself and its concerns, a reflexive research process. The reflexivity, sensitivity, and transdisciplinarity of the complex object Health can thus be incorporated into what is still a relatively atypical approach, even within a paradigm of complexity. In this tentative process of constructing a General Theory of Health, it will certainly be necessary to transcend the disciplinary borders between the so-called "natural" and human sciences in health. Thus, perhaps the health sciences, both human and natural, may actually deserve the title of Life Sciences.

Acknowledgments

Study funded by the Brazilian National Research Council (CNPq) with Fellowship Award 520.573/95-1 and Grant 463.855/00-0. Maria Thereza Ávila and Fernanda Tourinho Peres contributed extensively to the literature review, Jairnilson Paim, Maurício Lima Barreto, and Lígia Vieira da Silva collaborated with critical reading of the various versions of the paper, and Denise Coutinho proofread style and wording.

References

ABED, L. C., 1993. La Enfermedad en la Historia. Córdoba: Dirección General de Publicaciones, Universidad Nacional de Córdoba.         

ALMEIDA FILHO, N., 1997. A Clínica e a Epidemiologia. 2^a Ed. Rio de Janeiro: APCE/ABRASCO.         

ALMEIDA FILHO, N., 2000. What does the word "health" mean? Cadernos de Saúde Pública, 16: 300-301.         

ALMEIDA FILHO, N., in press. O conceito de saúde: Ponto-cego da epidemiologia? Revista Brasileira de Epidemiologia.          

ALMEIDA FILHO, N.; CORIN, E. & BIBEAU, G., unpublished manuscript. Signs, Meanings and Practice in Mental Health: Part II. Methodological Application. The Bahia Study.         

AYRES, J. R., 1997. Sobre o Risco - Para Compreender a Epidemiologia. São Paulo: Editora Hucitec.         

BERLINGUER, G., 1988. A Doença. São Paulo: Centro Brasileiro de Estudos de Saúde/Editora Hucitec.         

BIBEAU, G. A., 1988. Step toward thick thinking: From webs of significance to connections across dimensions. Medical Anthropology Quarterly, 2:402-416.         

BIBEAU, G., 1994. Hay una enfermedad en las Américas? Otro camino de la antropología médica para nuestro tiempo. In: Cultura y salud en la construcción de las Américas, VI Congreso Colombiano de Antropología, Anales, pp. 44-70, Bogotá: Instituto Colombiano de Cultura, Universidad de los Andes.         

BIBEAU, G. & CORIN, E., 1994. Culturaliser l'épidémiologie psychiatrique. Les systèmes de signes, de sens et d'action en santé mentale. In: Marc-Adélard Tremblay ou la construction de l'anthropologie québécoise (P. Charest, F. Trudel & Y. Breton, dir.), Quebec: Presses de l'Université Laval.         

BIBEAU, G. & CORIN, E., 1995. From submission to the text to interpretative violence. In: Beyond Textuality. Asceticism and Violence in Anthropological Interpretation. Approaches to Semiotics Series (G. Bibeau & E. Corin, ed.), pp. 3-54, Berlin: Mouton de Gruyter.         

BOORSE, C., 1975. On the distinction between disease and illness. Philosophy and Public Affairs, 5:49-68.         

BOORSE, C., 1977. Health as a theoretical concept. Philosophy of Science, 44:542-573.         

BOORSE, C., 1997. A rebuttal on health. In: What is Disease? (J. Humber & R. Almeder, ed.), pp. 1-134, New Jersey: Humana Press.         

CANGUILHEM, G., 1966. Nouvelles Réflexions sur le Normal et le Pathologique. Paris: P.U.F.         

CANGUILHEM, G., 1978. O Normal e o Patológico. Rio de Janeiro: Forense Universitária.         

CANGUILHEM, G., 1990. La Santè: Concept Vulgaire et Question Philosophique. Toulouse: Sables.         

CAPONI, S., 1997. Georges Canguilhem y el estatuto epistemológico del concepto de salud. Physis - Revista de Saúde Coletiva, 4:287-307.         

CHAVES, M., 1972. Saúde & Sistemas. Rio de Janeiro: Editora da Fundação Getúlio Vargas.         

CLAVREUL, J., 1980. A Ordem Médica. São Paulo: Brasiliense.         

CLOUSER, K.; CULVER, C. & GERT, B., 1997. Malady. In: What is Disease? (J. Humber & R. Almeder, ed.), pp. 175-217, New Jersey: Humana Press.         

CORIN, E., 1995. The social and cultural matrix of health and disease. In: Why are some People Healthy and others Not? The Determinants of Health of Populations (R. G. Evans, M. L. Barer & R. Marmor, ed.), pp. 93-132, Hawthorn: Aldine de Gruyter.         

CORIN, E.; BIBEAU, G. & UCHÔA, E., 1993. Éléments d'une sémiologie anthropologique des troubles psychiques chez les Bambara, Soninké et Bwa du Mali. Anthropologie et Sociétés, 17:125-156.         

CORIN, E. & LAUZON, G., 1992. Positive withdrawal and the quest for meaning: The reconstruction of experience among schizophrenics. Psychiatry, 55: 266-281.         

COSTA, N., 1989. Ensaio sobre os Fundamentos da Lógica. São Paulo: Editora Hucitec/EDUSP.         

CULVER, C. & GERT, B., 1982. Philosophy in Medicine. Oxford: Oxford University Press.         

CZERESNIA, D., 1999. The concept of health and the difference between prevention and promotion. Cadernos de Saúde Pública, 15:701-709.         

ENGELHARDT, T., 1975. The concepts of health and disease. In: Evaluation and Explanation in the Biomedical Sciences (T. Englehardt & S. F. Spicker, ed.), pp. 125-141, Dordrecht: D. Reidel Publishing.         

FIELD, D., 1976. The social definition of illness. In: An Introduction to Medical Sociology (D. Tuckett, ed.), pp. 334-366, London: Tavistock.         

FOUCAULT, M., 1963. Naissance de la Clinique: Une Archéologie du Régard Médical. Paris: P.U.F.         

FOUCAULT, M., 1976. Mental Illness & Psychology. New York: Harper & Row.         

FULFORD, K. W. M., 1994. Closet logics: Hidden conceptual elements in the DSM and ICD classifications of mental disorders. In: Philosophical Perspectives on Psychiatric Diagnostic Classification (J. Sadler, M. Schartz & O. Wiggens, ed.), pp. 211-232, Baltimore: Johns Hopkins University Press.         

GADAMER, H. G., 1993. The Enigma of Health. Berkeley: Stanford University Press.         

GONÇALVES, R. B. M., 1992. Práticas de Saúde: Processo de Trabalho e Necessidades. Cadernos CEFOR - Textos 1. São Paulo: Centro de Formação dos Trabalhadores da Saúde.         

GOOD, B., 1994. Medicine, Rationality, and Experience. An Anthropological Perspective. New York: Cambridge University Press.         

GOOD, B. & GOOD, M. J., 1980. The meaning of symptoms: A cultural hermeneutic model for clinical practice. In: The Relevance of Social Science for Medicine (L. Eisenberg & A. Kleinman, ed.), pp. 165-196, Dordrecht: D. Reidel Publishing.         

GOOD, B. & GOOD, M. J., 1982. Toward a meaning-centered analysis of popular illness categories: 'Fright Illness' and 'Heart Distress' in Iran. In: Cultural Conceptions of Mental Health and Therapy (A. J. Marsella & G. M. White, ed.), pp. 141-166, Dordrecht: D. Reidel Publishing.         

HELLER, A., 1986. La Teoria de las Necessidades en Marx. Barcelona: Península.         

HESSLOW, G., 1993. Do we need a concept of disease? Theoretical Medicine, 14:1-14.         

HUMBER, J. & ALMEDER, R. (ed.), 1997. What is Disease? New Jersey: Humana Press.         

INECOM, 1993. The International Network for Cultural Epidemiology and Community Mental Health. Montreal: World Health Organization Collaborating Centre for Research and Training in Mental Health.         

KAPLAN, B.; CASSEL, J. & GORE, S., 1977. Social support and health. Medical Care, 15:47-58.         

KAWACHI, I., 1999. Social capital and community effects on population and individual health. Annals of the New York Academy of Sciences, 896:120-130.         

KAWACHI, I.; KENNEDY, B. P. & GLASS, R., 1999. Social capital and self-rated health: A contextual analysis. American Journal of Public Health, 89:1187-1193.         

KLEINMAN, A., 1980. Patients and Healers in the Context of Culture. Berkeley: University of California Press.         

KLEINMAN, A., 1986. Concepts and a model for the comparison of medical systems as cultural systems. In: Concepts of Health, Illness and Disease. A Comparative Perspective (C. Currer & M. Stacey, ed.), pp. 29-47, Oxford: Berg Publishers.         

KLEINMAN, A., 1988. The Illness Narratives: Suffering, Healing & The Human Condition. New York: Basic Books.         

KLEINMAN, A., 1992. Local worlds of suffering: An interpersonal focus for ethnographies of illness experience. Qualitative Health Research, 2:127-134.         

KLEINMAN, A.; EINSENBERG, L. & GOOD, B., 1978. Culture, illness, and care. Clinical lessons from anthropologic and cross-cultural research. Annals of Internal Medicine, 88:251-258.         

LAKOFF, G., 1993. Women, Fire, and Dangerous Things. Berkeley: University of California Press.         

LEVINE, S., 1995. The meanings of health, illness and quality of life. In: Quality of Life and Health. Concepts, Methods, and Applications (I. Guggenmoos-Holzmann, K. Bloomfield, H. Brenner & U. Flick, ed.), pp. 7-14, Berlin: Blackwell Wissenschafts-Verlag.         

MASSÉ, R., 1995. Culture et Santé Publique. Montreal: Gaetan Morin.         

MURPHY, E., 1965. The Logic of Medicine. Baltimore: Johns Hopkins University Press.         

NOACK, H., 1987. Concepts of health and health promotion. In: Measurement in Health Promotion and Protection (T. Abelin, Z. Brzezinski & V. Carstairs, ed.), WHO Regional Publications, European Series 22, pp. 5-28, Copenhagen: World Health Organization.         

NORDENFELD, L., 1987. On the Nature of Health. Dordrecht: D. Reidel Publishing.         

NORDENFELD, L., 1993. Quality of Life, Health and Happiness. Aldershot: Avebury.         

OXFORD DICTIONARY, 1968. Oxford English Dictionary. New York: Oxford University Press.         

PAIM, J., 1996. A situação de saúde no Brasil e os modelos assistenciais. In: IV Encontro Especial da Sociedade Brasileira para o Progresso da Ciência (SBPC), Anais, pp. 249-253. Feira de Santana: SBPC.         

PARSONS, T., 1951. The Social System. New York: Free Press.         

PARSONS, T., 1964. Social Structure and Personality. New York: Free Press.         

PARSONS, T., 1975. The sick role and the role of the physician reconsidered. MMFQ/Health Sociology, 53:257-278.         

PARSONS, T., 1978. Action Theory and Human Condition. New York: Free Press.         

PEREZ-TAMAYO, R., 1988. El Concepto de Enfermedad. México, DF: Fondo de Cultura Económica.         

PÖRN, I., 1984. An equilibrium model of health. In: Health, Disease, and Causal Explanations in Medicine (L. Nordenfeld & J. Odelstad, ed.), pp. 225-238, Dordrecht: D. Reidel Publishing.         

PÖRN, I., 1993. Health and adaptedness. Theoretical Medicine, 14:295-303.         

SAMAJA, J., 1994. Epistemología y Metodología. Buenos Aires: EUDEBA.         

SAMAJA, J., 1997. Fundamentos Epistemológicos de las Ciencias de la Salud. Doctoral Thesis, Rio de Janeiro: Escola Nacional de Saúde Pública/Fundação Oswaldo Cruz.         

SUSSER, M., 1973. Causal Thinking in the Health Sciences. New York: Oxford University Press.         

TEMKIN, O., 1963. The scientific approach to disease: Specific entity and individual sickness. In: Scientific Change: Historical Studies in the Intellectual, Social, and Technical Conditions for Scientific Discovery (A. Crombie, ed.), pp. 629-647, New York: Basic Books.         

YOUNG, A., 1980. The discourse on stress and the reproduction of conventional knowledge. Social Science and Medicine, 148:133-146.         

YOUNG, A., 1982. The anthropologies of illness and sickness. Annual Reviews of Anthropology, 11:257-285.