Cultural interpretations of contagion


date Fondazione Angelo Celli per una cultura della salute, via del Giochetto 6, 06126 Perugia, Italia. E-mail: and since smallpox was officially eradicated in 1978, reference is presumably made to some exanthematic diseases such as chicken-pox. First of all it must be pointed out that the word indicates a benign disease. In fact, the Alladian have always feared nnowie akra-kra so much that the fact that the word means ‘benign disease’ indicates the wish to reduce its malefic power, and to prove that one is not afraid. This attitude towards the disease applies also at individual level: if for example a person is afraid to visit someone suffering from nnowie akra-kra, his fear can represent a reason why he will fall ill. But while at an empirical level experience shows that these infectious diseases can be transmitted through direct contact with infected persons and that therefore a certain behaviour must be adopted to keep the disease from spreading, this behaviour itself can be viewed as an expression of fear of the disease and be sufficient reason for contracting it. People may find themselves caught between contradictory interpretations that lead to diametrically opposed courses of action. Then there is the problem of individual transgression of the rules, of subversion of the established order; a key issue particularly within family contexts, in the relationships between father and son, husband and wife, but also between different generations, if the age groups' structure and the relationship between them are questioned. One example is niambure, a disease caused by incest between two people of the same maternal lineage (a man with his sister or cousin on the mother's side). For the disease to appear both sides must be willing to commit an infraction and to be aware that they are breaking a rule. It is not an incestuous relationship of the kind described by Héritier (1994), the specificity lies in the fact that to become infected, one must willingly break the rules. Pisa develops in the husband after the wife has committed adultery with another man and it involves the rules regulating sexuality and social behaviour. What is important to emphasize is that the disease can then be transmitted to other people through different forms of contagion involving the relationship between pure and impure, the concept of the body, of the individual, the relationship between men and women. The extreme aetiological variability requires an analysis of the forms of transmission of the disease case by case, taking into account the different circumstances.


Anthropological research in recent years has examined how single diseases such as Aids, tuberculosis, measles, malaria and leprosy are conceptualized by laypersons in non-Western societies. But how is disease transmission itself interpreted in other cultures? Data from ethnographical studies in Côte d'Ivoire and the Afro-Brazilian culture in Bahia, Brazil show that the interpretations of contagion and preventive practices cut across society involving five main relationships: empirical and analogical thinking, symbolic factors and social organization, the concept of person and body elements, natural and supernatural powers and individual and contextual factors. There is not a general theory, such as Pasteur's theory of germs. Instead,contagion presents itself as a transversal, multidimensional concept crossing and interconnecting society and culture. Public health programmes aimed at controlling infectious diseases need first to understand how contagion is conceptualized by laypersons, the extent to which diseases are considered infectious and the relation between perceptions and preventive practices. This would help in implementing infectious disease control programmes within local contexts based on meaningful community participation.


In 1982 I began to wonder what theories peoples and cultures outside Western society had on the transmissibility of diseases. (The terms contagion and transmissibility will be used synonymously here, although in some cases they can be different concepts.) At the time, I was developing health education, prevention and promotion programmes in Peru with the Quechua populations in the Andean region. Several projects involved the construction of latrines, but almost all of them, built in collaboration with the local communities, were being used as chicken hutches or for some other purpose. This experience prompted several questions such as: What is the point, in other cultures, of building latrines when the concept of oral-faecal transmission does not exist? Does the concept of contagion exist in non-Western cultures? If so, what are the causes, the different forms of person-to-person transmission, the preventive and therapeutic practices? What sort of relationship is there between these interpretations and a society's internal organization, the concept of the body, of life and death?

Very few attempts have been made to answer these general questions, particularly in the anthropological field. Indeed, most published works on infectious diseases in African societies examined the question from different standpoints such as public health, clinical health, epidemiology, social history (Turshen 1977; Janzen 1978; Dawson 1979; Feierman 1979; Ford 1979; M'Bokolo 1984), medical ecology (Wiesenfeld 1967; Hughes & Hunter 1970), history of diseases (Ackerknecht 1965; Patterson 1984) and history of colonial medicine (Pluchon 1985). Data were strikingly limited, particularly in anthropology (Dupire 1985; Perrin 1985; [Fainzang S (1986) ]Fainzang 1986, 1996; Nichter 1989; Bibeau 1996).

Anthropological research had looked at the subject from specific points of view: aetiology, diagnosis, therapy and the role of traditional healers and in other cases it had focused on single infectious diseases, in connection with health control and prevention programmes. At the end of the 1970s, the implementation of programmes aimed at controlling some of the diseases with the greatest public health impact had led many anthropologists to analyse the cultural and social aspects of single illnesses, particularly perception and behaviour in connection with infectious diseases that could be controlled through immunization (Heggenhougen & Clements 1987), diarrhoea (Coreil & Mull 1988; Kendall 1989), acute respiratory diseases, tuberculosis (Nichter 1994) and Aids. Most of these studies were part of the general medical anthropological research applied to international health programmes (Coreil & Mull 1990). A few studies tried to examine the problem of contagion within a given community, focusing on collective interpretations and practices, in terms of both prevention and therapy.

Studies first in Côte d'Ivoire and then in Salvador in Bahia (Brazil) revealed that there are some common principles in the way disease transmission is interpreted among the Alladian in Côte d'Ivoire and the Afro-Brazilian cultures in Bahia, but there are also some fundamental differences. The Alladian live in 14 villages on on a strip of land of roughly 60 km between the sea and the Ebrié lagoon on the Atlantic coast of the Côte d'Ivoire, about 50 km south-west of Abidjan. They once lived from fishing but this activity no longer plays a key role in the family economy and has been mostly replaced by the cultivation of coconuts. The Alladian kinship is matrilineal, with patrilocal residence. Several matrilineages (ecoko) joined by kinship form a matriclan (eme) (Augé 1969). Another important element in the social structure is the age group (esuba), which relates individuals according to their age, regardless of their different lineages.

In my studies I have tried to prove that the concepts relative to the contagion of disease are structured on a few main principles: empiricism and analogical thinking; symbolic classifications and the system of representations; the concept of the person and of the body; intrapersonal relationships and social organization; the relationship with supernatural powers; and individual factors and external circumstances ([8][Caprara A (1994) ]Caprara 1991, 1994; 1996a). In Western culture, infectiousness of diseases is explained by the theory of germs, and preventive or therapeutic practices are generally developed on this basis. Among the Côte d'Ivoire Alladian and the Afro-Brazilian cultures in Bahia, contagion is connected with everyday behaviour and events involving social dynamics rather than a specific single theory. Contagion therefore represents a transversal category linking disease on the one hand to the community's social organization, and on the other hand to its cultural values. It is a multifaceted concept, permitting individual and circumstantial factors to modify the rules of transmission. Fear of falling victim to a disease, a desire to break the rules, envy for someone, individual predisposition to a certain disease are all factors that facilitate contagion. It is precisely the concept of contagion and connected practices that lie at the basis of the differences, contradictions, conflicts and inconsistencies arising in traditional societies when the Western biomedical model is introduced.

Problems in interpretation

There are several initial problems which emerge as soon as this problem is examined. The first is of a semantic nature: for example, among Alladian, kotõ is the word used to indicate the transmissibility of a disease from one person to another. The meaning expressed by this word is connected with the action of closing or tying something up. Buckling a belt or closing a door are actions expressed by this word. In connection with disease, kotõ expresses the idea of a link, of a relationship established between two or more people, as in the case of pisa. Pisa is, par excellence, the disease of guilt, the illness that expresses breach of a social rule. Adultery is often considered to be its cause, but it can also be brought about by many other infringements of social rules. It is a serious disease which often leads to death; it can infect both men and women, causing what the Alladian describe as: ‘a real and proper destruction of the person, which makes men impotent and women barren’. The illness spreads through several mechanisms of transmission and develops when a sort of circuit is formed between three individuals, two men and a woman. In the case of adultery committed by the wife, the disease makes its appearance in the husband when he comes into direct or indirect contact, by chance or voluntarily, with his wife's lover. This can happen even many years after adultery was committed. The disease will appear when the two men meet and touch. The lover therefore must avoid all contact with the husband, even helping him in difficulties, to avoid creating the paradoxical situation that could turn his help into a gift of death. The risk of contact is frequent since people often shake hands in greeting each other and introduce themselves on occasions such as funerals, normally attended by the entire population. The disease thus becomes the expression, the sign of transgression of a basic rule established by the community. It clearly indicates the existing bonds between the forms of human communication, social rules and disease. Thus the words kotõ and pisa necessitate an examination of the problem of contagion from the viewpoint of social relations between individuals and their role within the community.

The Bété, another major ethnic group in Côte d’Ivoire, a patrilineal, patri-local and exogamous society (Caprara et al. 1993), call the transmission of an illness plenagou, ‘a disease which goes from one point to another’. Its interpretation is based on the relationship between four categories of thought: symbolic, empirical, social and supernatural. The Bété distinguish some elements through the juxtaposition of pure and impure. Contact with impure elements causes infection: if a person walks or urinates where an ill person has urinated, he may get the same disease, because urine, menstrual blood and saliva are considered impure elements. An empirical explanation of contagion is also admitted. Disease transmission may occur through direct contact with the patient, e.g. if a person helps a patient affected by a disease considered infectious (Caprara et al. 1993). Some diseases have a strong social implication, for example those transmitted by inheritance: leprosy and tuberculosis are perceived as hereditary ‘family diseases’ because they are recognized as related to a specific family. Thus epilepsy too is regarded as infectious.

In the Afro-Brazilian cultures of Bahia, infectious diseases are defined as doenças que pegam; the verb pegar expresses the idea of glue, sticking a disease on a person. It is a question therefore of understanding who transmits the disease to whom and in what specific context. The researcher must identify and analyse the words used to express contagion. In this way, a semantic network can be constructed which links infectious diseases with symbols, communication rules and body concepts.

The second problem is of a nosographical nature and involves the many differences in classification between Western and non-Western categories. For example, the Alladian nnowie akra-kra has been translated as smallpox, ebrouakpe as haemorroids, pisa as tuberculosis (Duponchel 1974). These translations are misleading. Pisa, described in literature as equal to tuberculosis, manifests very different symptoms. A description by an Alladian healer confirms this: ‘If you are wounded and your blood flows incessantly, you have pisa; if you have a scar which continues to bleed, everyone will know what disease you have contracted, you have pisa’.

The diagnosis is indeed made more on the basis of the situation rather than the symptoms as such. Therefore it is necessary for an abnormal situation to arise (complications linked sometimes to different pathologies) if pisa is to be diagnosed. Blood is still often at the centre of clinical manifestations and, as healers are apt to emphasise, ‘pisa is a blood disease’. A close link is established between the situations indicating the disease and the flow of blood. Thus rather than seeking a direct connection between nosographical categories, it would be preferable to understand the semantic networks that link various pathologies (Good 1977; Bibeau 1981; Nichter 1994).

The third problem involves the transmissibility of diseases: what is considered contagious in one society may not be in another. For example, epilepsy is considered infectious in both the African and Afro-Brazilian cultures (Beiser 1985; Awaritefe 1989; Caprara 1996b). Popular Bahian culture views epilepsy as contagious since transmission is possible through the patient's saliva, in this case considered a symbol of impurity. The elements of stigma and contagion which can produce social isolation do not only involve proxemics – do not touch a sick person's body, his/her things or places s/he may have touched – but may take on aspects of rejection and marginalization which have as yet been only superficially explored. The west-African and the Afro-Brazilian cultures have in common the link between pure and impure, the concept of body fluids as a source of disease. Illnesses considered infectious by Western medical science are not necessarily viewed as such in other contexts, and it is therefore fundamental to clarify these differences in classification from the beginning, since they may produce deep divergencies, misunderstandings and paradoxes later, when health programmes are implemented.

Cultural interpretations

Both in some ethnic groups in the Côte d'Ivoire and among the Afro-Brazilian cultures, disease transmission is considered within a framework of relationships between analogical thinking and empirical experience, symbolic classifications and social organization and individuals and body conceptions. Regarding other relationships, there is a connection between symbolic elements, social organization and the concept of the body and person, e.g. the concepts of pure and impure clearly define how diseases can be transmitted through impure body fluids (Douglas 1970).

From analysing the data collected through interviews with key people in the community and from case histories, it emerged that empirical thinking involves the way in which the Alladian interpret the possible causes of a disease. If, for example, the disease develops in an individual after he has come into contact with a sick person, ‘empirical’ interpretation refers to the direct transfer of the disease from one individual to another. If this situation does not arise, the cause is based on other forms of interpretation. It is therefore above all a practical view, based on daily experience, that makes it possible to understand the cause and transmission of a disease.

This practical interpretation structure is accompanied by analogical thinking. Often transmissibility is interpreted through categories linked to each other by a similarity in form, colour, substance or proximity. For example, for the Alladian, one of the better known infectious diseases is called ebrouakpe, a word indicating both the veins and nerves in the body structure and the roots of trees. The analogical relationship implied by the word's semantics establishes a connection between certain elements of the human body and plants. Several authors have stressed the importance of this form of interpretation in understanding reality in Western (Foucault 1966; Frazer 1973) and other cultures ([5]Bibeau 1981, 1992; Nichter 1989). In Bahia empirical and analogical thinking are linked and supply a key to interpretation, as in the example of a case of chicken-pox, described by a head priestess (Mãe de Santo): ‘The son of a friend was ill with chicken-pox and a niece of mine fell ill shortly thereafter. She had entered the other child's house and then left it. A few days later she had a temperature and immediately her body became covered with spots’ (translation by the author).

In addition to empirical thinking, analogical thinking provides an important model of interpretation in the Bahian culture: the wounds of Omolu, an important divinity in the Candomblé religious cult connected with all infectious diseases and skin pathologies are closely linked with skin lesions on some sick persons and popcorn. Analogical thinking provides the means to establish how certain elements affect others through their conceptual similarity.

Practical implications for public health programmes

At a practical level, the study of the various interpretations of contagion in different situations is important for the implementation of health education and promotion programmes. While the epidemiology of many infectious diseases is well-known, the socio-cultural aspects are mostly unknown. An ethnographically orientated approach should have the objective of studying infectious diseases and preventive measures against them starting from interpretations developed by the community. This sort of approach goes beyond a simple adaptation of health programmes to a community's cultural context, and benefits of real contact with the population's learning processes and community texture. How can one expect to change a population's behaviour regarding prevention or hygiene when one is dealing with the coexistence of different thinking patterns? This question does not cause us to underestimate the social and economic factors linked to the spread of disease; rather it forces us to extend our area of analysis and introduce greater complexity.

Indeed there is a great deal of controversy over the question of changes within a culture. On the one hand, there are those who claim that behaviour can be somewhat modified even if a given group does not understand the reasons behind these changes, while on the other hand it is felt that the population must understand the rationale behind actions. In most developing countries, the introduction of Western models of infectious disease control has been imposed to replace a different traditional way of thinking, linked to the context of origin. If a model proves to be effective in the short term, it will be more easily accepted by a community, but if it is to be fully integrated, it must also be understood (Caprara 1991). I believe that it is necessary to understand how a community interprets certain diseases and what practices are adopted to prevent them in order to develop educational processes that produce meaningful community involvement (Nichter 1990).

As a starting point the hidden meanings behind a community's actions must be elucidated to be able to implement health programmes that will make sense in local cultural contexts. For example, if AIDS prevention methods are to be adopted, the basic concepts concerning the transmissibility of the disease must be clear and accepted by the community as a whole. This apparently simple matter becomes very complex when one analyses the categories on which the concept of transmissibility of a disease is structured in a traditional context (Caprara et al. 1993). The Bété interpretations and preventive practices necessitated to reflect on how infectious diseases are considered, and which traditional practices are adopted to stop transmission.


In my view we need to deal with this matter in its complexity, trying to understand the rules, strategies and traditional practices, concentrating not so much on individual diseases but rather on the problem of contagion in general. Only then will it be possible to analyse the traditional categories linked to single diseases, and in this sense include them in a broader context. This should be the focus of research on the processes of transformation, re-definition, and changes taking place in many traditional societies with regard to problems involving health and disease. The question of the transmissibility of diseases, of traditional prevention and control methods makes it possible to concentrate on the cognitive and practical problems that develop in a culture when different medical categories coexist. It leads to a more careful analysis of the relationship between empirical and analogical thinking, of the social organization and cultural values, of the concepts of the body and the individual.


I would like to thank Professor Gilles Bibeau, University of Montreal, Professor Naomar de Almeida Filho, Federal University of Bahia, Professor Tullio Seppilli, University of Perugia and Professor Vivaldo da Costa Lima CEAO, Bahia, for their comments and suggestions on earlier drafts. I am particularly grateful to Dr Silvia Mamede, Director of the School of Public Health of Ceará, Lucy Wood and Giovanna Simmons for assistance with this research. The project was carried out in Côte d'Ivoire from July 90 to September 92 and then in Brazil, at Salvador (Bahia). The project in Brazil is run in collaboration with the School of Public Health of Ceará, and funded by a grant from the North-East Project 2/World Bank, Brazilian Ministry of Health. A summary of this research was presented at the XX Brazilian Conference of Anthropology, Bahia, Brasil, 14–18 abril 1996.