Contents - Previous - Next

This is the old United Nations University website. Visit the new site at

Perception of physiological effects and food classification

Individual foods or methods of food preparation are also classified and ranked according to their perceived physiological effects. The physiological sensation of satiety is a first factor. For example, the Bemba of Zambia classify the different grains they prepare as porridges according to "fillingness." They aim toward a "turgid" feeling in the stomach and further rank the different grains - millet is ranked higher than maize or sorghum - in terms of the amount of time porridges made from them will supply this feeling and suppress hunger pains (Richards, 1939). Hondurans have been reported to prefer maize over sorghum for the same reason tortillas of the former are described as "more filling" (DeWalt, 1983). It remains to be determined whether their perceptions have some basis in the nutrient composition of the grains whether nutrient (protein, fat, or caloric) density, gross fibre or bulk, or some other qualities. Such perceptions, however, will affect what and how much of a particular staple people eat and may be important either for trying to change diets to a more productive staple grain or for trying simply to increase the intake of the existing grain. In recording such food preferences, one should also, if possible, see if consumption of one versus the other product conforms to the stated verbal expectations of relative fillingness.

Protein may be an additional factor. Hunting and gathering cultures put a high value on "meat" and claim that, without it, they are "hungry" no matter how much vegetable food they have ingested (Lee, 1968; Holmberg, 1950). Horticultural populations who want protein, fat, and other nutrients may also continue to forage to provide a variety of meats, fish, and other foods for their diets.

Johnson and Baksh (1983), working among the Camana and Shimaa of the Peruvian Amazon, compared the relative productivity of fishing and hunting/collecting in terms of kilograms per hour, caloric efficiency, and protein efficiency. They argued that the wide variety of wild foods in the diet, each present in small amounts, met a number of nutrient and 'psychological" needs. Furthermore, they suggested a group would tend to exploit more and more a "broad spectrum" of resources as populations of dominant wild resources decreased in size as a result of human activities (Flannery, 1973). Desire for these food supplies may cause groups to move temporarily, or, finally, permanently, to locations where they are more abundant.

Agricultural populations may also be "hungry for meat," though whether it is the actual ingestion of the meat or the anticipation of the feasting, the heightened activity, and the feelings of exhilaration that may accompany meat consumption is open to question. Richards (1939), for example, found that the Bemba increased their activity and changed their mood in anticipation of eating meat, even before they had the chance to enjoy its benefits. Others have suggested that the emotional and physiological cravings for meat reported in the ethnographic literature might be more accurately a function of the salt or fat content of the meat (Hayden, 1981; Jochim, 1981). Jochim (1981, pp 80-83) suggests that diets for particular cultures and for crosscultural comparison be compared not in terms of productivity or meat versus vegetable foods, but by the proportions of fat and protein in the particular foods and the diets as a whole. In this manner, one might arrive at a fat or a protein factor in food rankings that would go a long way to providing an additional nutritional component of taste preferences. Values for protein and fat can be retrieved from standard food composition tables or tapes, and then plotted to show relative rankings. In this way, one can translate from cultural taste and physiological values to nutritional contents, which can be compared cross-culturally.

In addition, fat and salt, along with other condiments, are recognized as physiologically necessary by most peoples subsisting on a predominantly cereal grain, tuber, or plantain diet. The Bemba, for example, consume grain porridges with a relish of animal or vegetable protein, at a ratio of three parts grain to one part relish. If the female head of the household has not had the energy to gather the relish ingredients from the bush, she might not prepare the grain, since, in her cultural view, one cannot get the porridge down without the relish. One might make a similar case for the consumption of tortillas with chili peppers in Mesoamerica. Without pungent spices, natives claim that they have no "appetite."

Cultural dietary practices thus take into account the need for spices to help people ingest sufficient quantities of their bulky, starch staples to meet energy needs. More detailed cultural interviews and observations on the typical rice, maize, wheat, or tuber diets may contribute to scientific as well as cultural understandings of quantitative food intake, where intake is not absolutely limited by supply. Detailed study may also provide a better understanding of how the different combinations of staples with relishes and condiments may be related to growth patterns; that is, different outcomes of weight and height for age seem to occur under different dietary regimes. In additions, such dietary combinations may constitute factors impinging on hunger, appetite, satiety, and body composition. These interrelations among diet, food intake, and growth patterns merit detailed observation and interpretation.

Some spices may provide other physiological and emotional responses which people learn to classify as desirable, e.g. chili peppers increase the flow of saliva and intestinal activity, both of which aid in the digestion of a maize or other high bulk diet (Rozin and Schiller, 1980). Other physiological effects that become incorporated into cultural dietary preferences include: (1) the stimulating effects of caffeine in coffee, tea, chocolate, and cola drinks; (2) the sudden rush of energy that people experience after consuming heavily sugared beverages (Messer, 1986); and (3) the effects of appetite stimulation or dulling of certain "quasi-food" medicinals and tonics' like marijuana, coca leaves, ch 'at (Catha edulis, a leaf stimulant, chewed commonly in Ethiopia; Simoons, 1960), and alcohol-based substances. Ingredients of bitter taste are often included as elements of herbal brews and tonics prepared to stimulate appetite in many parts of the world, since "bitterness" is recognized as characteristic of certain underlying chemical constituents like tannins, which do have a perceptible physiological effect on gastric secretions (Etkin and Ross, 1982).

Negative short-term physiological effects may, conversely, form the basis of food avoidances. Fischer, Fischer, and Mahoney (1977), in their analysis of foods avoided by a group of Caroline Islanders, found that most of the foods that were tabooed (ritually prohibited) were potential allergens, such as shellfish and eels. Furthermore, "punishments" for breaking food taboos, whether purposefully or accidentally, corresponded to the symptoms of allergic reaction. They included chastisements that resulted in skin eruptions, skin sores, swelling, and shortness of breath. Within this culture, violations of prohibitions against consumption of the "runts" of species also existed. Consumption of "runts," which would be less likely than allergens to produce such adverse physiological effects, was supposed to be punished by a deformed birth in the future. Careful assessment of the possible negative physiological effects of taboos in other locations - arrived at by in-depth interviewing of native descriptions of anticipated consequences - may reveal similar rationales.

Culturally prescribed avoidances such as prohibitions on consuming shellfish and pork may also have their health rationale, although the cultural reasons for avoiding such items may go beyond the simple reasoning that they will "make you sick" (Douglas, 1966). Other adverse physiological reactions culturally encoded as food dislikes may be at least in part genetically based. For example, genetically based lactose intolerance may be at the root of milk avoidances in certain cultures (Simoons, 1973; Harrison, 1975), although the physiological argument still does not explain why certain lactose-intolerant populations, like the Chinese, do not like cultured milk products which they should be able to digest. Mediterranean populations tested to be lactose-intolerant have adjusted to the inherited inability to digest fresh milk by "predigesting" it into processed forms, like yogurts and cheese, which can be consumed without gastric distress. Thus, physiological and cultural factors must be considered in evaluating the evolution and distribution of food dislikes.

Ethnologists observing contemporary populations have noted how new foods are often classified as "good" or "bad" for people on the basis of their perceived physiological effects whether they are easily digested or seem to make people sick. Such idiosyncratic observations of particular foods may be incorporated into an existing cultural symbolic classification such as hot-cold, which systematizes health and dietary interactions (Cosminsky, 1977; Messer, 1981). What is still little understood, however, is how certain otherwise innocuous foods come to be classified as not only unacceptable, but as disgusting (Rozin and Fallon, 1981). For example, accidentally eating a "tabooed" food, like cobra, can make a native who respects that "taboo" vomit (see example from Holmberg, 1950).

Similarly, the psychological and cultural components of food cravings are not well understood. The factors permitting or encouraging consumption of "dangerous" foods, i.e. those that are tainted, toxic, or carcinogenic, are an additional area for investigation. Cultural descriptions, such as the "tingling" sensation that follows ingestion of the Japanese fugu fish, toxic if not prepared correctly, makes it highly desirable as a cultural delicacy, even though people know a small flaw in preparation will be lethal. Again, careful descriptions, from observation and interview, of what the local perceptions of sensory processes are, what sensations follow - in the short and long term - from the ingestion of particular foods, and how such sensory perceptions affect desirability and consumption of the particular items may help sort out some of these intriguing questions on the significance of the sensory dimension in the formation of food habits.

Cultural symbolic dimension

In addition to sensory characteristics, foods may also be classified according to a number of cultural factors, such as "hot-cold," "male-female," and "dangerous for pregnant women," which are culturally constructed from sensory data and other information. These may be the dimensions most prominent in the food proscriptions and prescriptions of nutritionally "vulnerable groups" such as infants, children, and pregnant and lactating women, and should, therefore, receive careful attention.

Hot-Cold and Other Binary Dimensions

Among the dimensions that singly or in combination have been used in different cultural contexts to classify food and to direct food intake are: hot-cold, wet-dry, male-female, heavy-light, yin-yang, pure-impure, clean-poison, and ripe-unripe. "Flavour," "sharpness," "itchiness," and "colour" are additional terms, less frequently encountered (Reichel-Dolmatoff, 1968; Ahern, 1975; Colson, 1976; Beck, 1969; Messer, 1981; Manderson, 1981). Such categories are often termed "symbolic" in that they may not refer to easily measurably or single "objective" qualities of food or other items, and also because such classifications often bring together a number of different socio-cultural domains, such as flora, fauna, medicine, health, ritual, and social relations. They reach their greatest elaboration in Eastern, particularly Indian and Chinese cultures, where they are part of a complex system of humoral medicine and philosophy. Their meaning and nutritional significance vary according to cultural context, but also individual inclination to "follow the rules."

Ideally, the best way to learn how such cultural rules operate is during extended periods of participant observation in a culture. In this manner, one can observe what kinds of questions people ask in cases of illness, to determine medicinal and nutritional dosing of the ill, and also under ordinary circumstances in the preparation of diet. This is usually not possible in shortterm survey work. Therefore, careful reading of the ethnographic and folkloric literature on the food habits of the locality or region, followed by careful interviewing of key informants to arrive at appropriate questions and categories for a particular culture, are necessary. During these interviews, one should try to ascertain background information - how cultural symbolic categories operate in the particular culture or relate concepts of food, medicine, health, and other areas. Then, one can query what types of foods and medicines are administered or avoided under particular illness conditions or during physiological changes, such as those due to pregnancy. Finally, one can add questions such as: "If these are the rules, do you follow them? Why?" making it clear to the respondent that in many situations people know the rules, but do not follow them, and that "no" is a perfectly acceptable response.

If one is interested in learning to what extent food classifications and rules for applying them to consumption are shared within cultures, it is also useful to design a list of sample foods and have people classify them. Then one can analyse uniformities in classification within a cultural group and determine (1) which items are largely classified in the same category; (2) which show significant variations; and (3) what the consequences are of such differences for designing, for example, an appropriate nutrition intervention. (On variable classifications, see Mazess, 1971; for discussion of other issues see Messer, 1981, especially references to relevant sections.)

Such preliminary information can then he used, if desired, to design questions about how such classifications are formed, how they are culturally transmitted, and how people within the culture communicate, if they are not operating on the basis of the same information. Messer (1981) found that while people shared the same general structure and rules for classification, they did not necessarily judge all items equivalently, given the differences in individual experience. Also, individuals differed in the extent to which they had acquired information and applied it to their own diets and health. Similar information collected in other areas can greatly aid in interpretation of food habits that might be beneficial or harmful to a particular population or subpopulation.

Among the symbolic classifications used in different parts of the world, hot cold classifications, which bring together the domains of food, health, and social relations, are among the most discussed in the literature, partly because they appear to span the Old World and the New. The hot-cold classifications are, therefore, useful for comparing and contrasting nutritional and medical ideas from different cultures and also for tracing the diffusion of foods, medicaments, and medical systems.

In brief, this system of classification is based on hot-cold values, which refer to an intrinsic quality rather than to temperature or spiciness and are ideally present within the human body in approximate balance. Too much heat or cold, caused by overconsumption of either hot or cold substances, overexertion, overheating the body, overexposure (to climatic or other heat or chills), an illness agent classified as giving rise to heat or cold within the body, or usually a combination of these factors produces an imbalance believed to result in illness. For example, an elderly Mexican peasant, "hot" from years of work in the hot sun, avoids overconsumption of chocolate, classified as hot, because he will find it difficult to digest, a cause-and-effect sequence he describes in terms of too much heat" (Messer, 1981).

Imbalances are corrected by the principle of opposites: the individual is carefully dosed with foods and/or medicines of the opposite quality to restore a healthy balance, and the person is also encouraged to avoid exposure to the offending quality. In the Mexican case just cited. the elderly farmer may correct an imbalance of too much heat with "cooling" herbal teas, e.g. tea of lemon grass, or a diet that avoids fats and spices classified as hot. In some cases, small quantities of substances with the quality that is in excess are introduced to prevent too severe a jolt to the system, which, it is believed, would further aggravate illness.

Hot-cold classifications of foods, medicine, and illnesses have been reported from many areas of the world, but they display great variability in how they are conceived and how they operate in local dietary and health practices (see Logan, 1977, and Messer, 1981, for reviews). It is best to consult the literature and then interview intensively in one's particular ethnographic location in order to find out the specific rules for classification and application, particularly if one is interested in introducing new foods or medicines (see case-studies by Harwood, 1971: Golpades et al., 1975; and Cosminsky, 1977).

Additional folkloric factors (Appendix 6), such as reproductive status, may also condition the consumption of certain foods in particular cultures (Manderson and Mathews, 1981). In each cultural case, however, individual knowledge and use of hot-cold and other cultural information must be considered to understand if they have significance for actual dietary constructions and, consequently, nutritional significance.

Health Factors

Culturally relative health beliefs also affect food choices. These beliefs include concepts of "safe" or "harmless" foods (foods that do not make one sick) and concepts of 'nutritious," "vitamin-rich," and "tonic" foods, which are thought to be positively good for health and wellbeing. These categories often subsume perceptions of physiological effects; foods that are good to eat and good for you are also those that seem to elicit good appetite and promote energetic well-being. Furthermore, in children they include foods that seem to foster growth and good health.

Within the "healthful" category, foods classified as "nourishing" and "vitamin-rich" are often analysed as "neutral" categories with respect to hot-cold, in the sense that they are considered to be generally beneficial and can be consumed in quantity without harm (Cosminsky, 1977). Such neutral foods often incorporate the basic staple of the diet, the nourishing value of which is also determined according to how filling the food is, i.e. how much it produces, and sustains, feelings of satiety. Certain Latin American cultures also judge the values of foods according to their "juiciness" (Young, 1981). "Dry" foods like cheese and dried fish are considered insufficient to sustain "life-force" or "strength" over long periods of time. More generally, the "cooked" food that defines a "meal" as opposed to a "snack" in some cultures, such as "cooked rice" in the Philippines, may be categorically defined as "nourishing" and, therefore, basic to health and nutrition. Information on these health dimensions of foods must be collected by careful interviewing of key informants prior to the design of nutrition interventions and should be checked, if possible, by ongoing in-house observations of actual consumption.

Additional questions for understanding the "health factors" in food choice are how concepts of spiritual and physical health may be tied to hot-cold, yin-yang, or other cultural symbolic qualities. For example, in Indian cultures, foods classified as "cool" are generally considered more healthful than those classified as "hot." Such beliefs are tied to more general cosmological conceptions of the physiological relationships between ingestion, digestion, and health.

Health food 'faddism" in Western culture is an additional example of how people use food classified as " healthful" to control both physical and emotional well-being. In this case, natural foods" (and natural food therapies) are seen as a kind of alternative spiritual and physical health maintenance system. In contrast to Indian or Chinese conceptual systems, Western health food faddists can point to no single classification system relating foods and health or any unified philosophy, and there is great individual variation among participants (Kandel and Pelto, 1979). Needless to add, "folk" and "scientific" ideas of "healthfulness" vary considerably.

People acquire ideas about the healthfulness of foods from diverse sources. Currently, both in Western countries and the third world, information about the nutritional value of foods is derived from advertising and medical personnel, as well as from accumulated cultural hearsay. Like the hot-cold classifications discussed above, nutritional and vitamin'' categories ordinarily may have little impact on diet. Under conditions of stress, however. particularly when one is weak because of illness, people may attempt to eat more vitamin-rich" food or take vitamin tonics (Messer, 1981). Again, interview, 24-hour recall, or, better, observation records, should be designed to examine aspects of cultural transmission.

Recent attempts to quantify the nutritional knowledge, beliefs, and attitudes that characterize food choices seem to indicate that the nutritional knowledge of the food provider is insufficient for predicting decision-making. (For additional methodologies see Sims, 1978; Worseley, 1980; Caliendo et al., 1976.) Even with adequate scientific nutritional knowledge, considerations of flavour and cost seem to take precedence over criteria of healthfulness (DeWalt and Pelto, 1976). Foods that are sweet and/or fatty are still preferred in many areas despite growing evidence that high intakes of either are unhealthful, and despite dissemination of nutritional information to this effect.

However, the reasons for such preferences are rarely reported, and careful interviewing might again reveal what taste or cultural factors people are selecting for in their food choices. Concentrated sugar consumption, for example, produces a rush of energy, which some people find pleasant and which motivates them for work. Fatty portions, in both developing and developed countries, may be considered more tasty and filling. High salt intakes may accompany cultural folk wisdom that certain foods, even water, are "unhealthful" without it (McCullough, 1973), particularly after exertion and sweating. Finally, people may also be disinclined to believe that there are negative health effects of particular diets unless they or close relatives have personally experienced the cause-and-effect relationships between particular foods and illness, such as diabetes, which is scientifically associated with high sugar intake.

People thus tend to select diet in large part on the basis of taste and appeal, rather than reputed nutritional value and the larger health consequences, unless confronted with the immediate spectre of supposed nutrition-related illness. To develop educational instruments showing intrinsic relationships between nutrition and disease, detailed examinations are still needed of the indicators by which people evaluate etiology in nutrition-related disease.

Age and Gender Factors in Food Selection and Restriction

Foods are also judged to be more or less appropriate for certain classes of individuals and for certain occasions. Again, a combination of interviews with key informants and observations of conditions and situations are the methods by which to record such data, which include the "rules" as well as inclinations to follow the rules. Such preliminary information can then be incorporated into a questionnaire on food habits.

Certain foods, for example, are judged to be especially good for or only edible by children. They are usually those that are observed to be pleasing to children; often they are bland (not overly spicy) and easy to digest. Since they do not make children sick, they are believed to foster growth. An interesting question is whether these are mainly food preparations not ordinarily consumed by adults. If a culture is used to preparing special foods for infants and young children, it may be more possible to target nutrition interventions, either special foods or education on special preparations of local foods, to them. Not all cultures, however, provide food prepared especially for children, a feature that often makes it difficult for the children to take in sufficient calories and may also create hardships for the nutrition planner trying to introduce special foods or food preparations for youngsters.

Restrictions on some foods for very young children are almost universal, although the particular foods and the rationales for withholding them differ from culture to culture. Such restrictions may have their basis in ideas of indigestibility; for example, whole grains of rice and beans are never served to children in certain cultures because people observe that whole grains pass undigested through the digestive tract into the stools. Or, hot cold classifications may influence ideas about which foods are deemed healthful for children; for example, young children in Mexico and elsewhere are classified as predominantly cool and their intakes of cold foods restricted, particularly if they are ill (Messer, 1981) (see Appendix 6.) While ordinarily they are allowed freely to imbibe "cold" liquids and sauces, these are restricted if the children exhibit signs of indigestion due to excess "cold" (e.g. loose, "green" atolls). In South-East Asia, young boys may have "cool" foods restricted, since their cool qualities are believed to interfere with growth and maturation. Because green, leafy vegetables and most fruits - major dietary sources of vitamin A - are classified as "cool," and because such foods are restricted, male children may suffer from vitamin A deficiency (Van Veen, 1971). In this case, the investigators sought and found other sources of vitamin A in the food environment that were not classified as cool and recommended that these be fed in larger quantities. In other parts of South-East Asia, such nutrition-health interventions are unnecessary as the food rules seem to be losing force (Manderson, 1981).

In addition to these types of food practices, more complex beliefs about and attitudes toward children at the household level - e.g. systematic favouring or neglect of children of one sex or at one position in the birth order - may affect infant and child nutrition and health. In such cases, rules for intra-household distribution of food as well as actual practices must be examined to see if there is systematic discrimination, or whether such practices vary according to other socio-economic and cultural factors.

Points in the female or male reproductive cycle, physiological stages, and advanced age may also be marked by food prescriptions and proscriptions in many cultures (Manderson and Mathews, 1981; Mathews and Manderson, 1981). Perceptions of body image or cultural ideals of altruistic behaviour, particularly for women, also affect general levels of intake at different points in the life-cycle. Indian women in the past were advised to restrict intake when pregnant to avoid difficult delivery; such denials also conformed to cultural expectations of personal sacrifice for the well-being of the family. Similarly, Indonesian women were known to restrict intake even when breast-feeding, in accord with cultural values that encouraged them to be selfsacrificing and slender. There may also be special food prescriptions and prohibitions for old age.

In each cultural case knowledge and attitudes and adherence to categories must be analysed, since there may be significant intra-cultural variation in beliefs and practices, and since what people say they do or do not eat may differ from what they actually consume. Such differences may he substantial even within cultures

Illness as a Factor in Food Selection and Restriction

Similarly illnesses, particularly those believed to be related to superhuman contacts-possession by spirits - may result in special food demands and privileges. More generally, culturally recognized "illness" is marked by alteration in eating behaviour as part of "social behaviours" in most of the world. Unwillingness to eat and lack of appetite or occasional overindulgence are signs of illness; restoration of "normal" appetite is the sign of a renewed health state. Background information on the culturally recognized illnesses and their possible treatments can be ascertained from the ethnographic literature of an area. Interviews should establish which illnesses elicit special dietary behaviours. In particular, it is important to understand general rules for handling respiratory and digestive disorders through diet. There is a growing literature on particular problems, such as diarrhoea-dehydration syndromes and methods of treatment and the possible pharmacological rationale for certain dietary-medicinal regimes (Etkin and Ross, 1982).

Gender as a Factor in Food Classification

Attributes of "male" versus "female" foods, their symbolism, and how they affect actual consumption are usually included in ethnographic reports, but additional interviews can help sort out which features of foods make them particularly good or bad for one or the other sex. A general methodology for assessing the significance of gender factors in food selection and nutrition is presented in Appendix 7. Symbolism in division of foods may follow closely the more general male-female divisions of labour and dominant-subordinate relationships in a society. O'Laughlin (1974) provides an excellent example of how to proceed in a symbolic-socioeconomic analysis of gender and dietary restrictions. First, she examined the complete domains of male and female activities and dominant-subordinate relationships. Next, she analysed male and female dietary restrictions (particularly restrictions on women from eating most animal protein sources and white refined-grain porridge) in symbolic terms, which, with adequate dietary intake data, could also be used for a nutritional analysis of women. A similar analysis of male and female foods was made for the Chagga in Tanzania (Swantz, 1975, cited in Pelto and Pelto, 1983).

Ritual and Economic Status as Factors in Food Selection

Ritual and economic status may also direct food intakes and avoidances. Members of a particular society mark totemic, caste, and religious group affiliations by sharing food avoidances, festival foods, and ordinary food preparations and consumption in common. Prestige factors also affect food choices and concepts of what is culturally "appropriate" for one of a given socio-economic status (or pretensions to that status) to consume. Thus, in Latin America, ethnographers report an increasing tendency for people to forgo ''wild" greens in favour of cultivated vegetables because consumption of wild foods is considered to be a sign of poverty (Messer, 1978; DeWalt et al., 1979). The problem of making low-cost foods, designed or aimed toward the poor, acceptable to them is also well-known (Gershoff, 1971). On the opposite end of the budgetary spectrum, relatively expensive foods may be consumed out of proportion with expenses for other food items because they are of high cultural value, as in the case of carbonated beverages in developing countries.

The various dimensions of personal status - biological, social, and economic - can positively or negatively affect nutrient intake. In large part these aspects of status supply the contexts in which other symbolic factors in food classification and selection operate, and thereby condition the nutritional impact of these other cultural food rules.

Contents - Previous - Next