Previous - Next
This is the old United Nations University website. Visit the new site at http://unu.edu
Introduction
Methods for studying cultural rules for food use
Research techniques
Social
units
Food choices: a process of many phases
Levels and units of analysis
A comprehensive interview approach to food
patterning
Conclusion
JUDITH G. GOODE
Temple University, Philadelphia, Pennsylvania, USA
This chapter focuses on the system of rules for organizing food intake and the predominant methodological approaches to the study of such cultural patterning. Cultural rules for food use are not limited to lists of avoidances (taboos) and preferences. These rules are more accurately characterized as group-shared systems of ideas about organizing food items into dishes and meals that ultimately influence the timing, order, and quantity of intake. Moreover. such ideas are constantly negotiated. They are communicated, reinforced and modified over time through interaction in meaningful social units (households and communities).
In most cultures there are rules for combining, segregating and sequencing foods. For example, the well-known Jewish rules of kashruth proscribe the mixing of flesh and dairy animal products in a single dish or food event. This is a segregation rule. Examples of pattern rules for combining foods can be found in many American dish patterns such as peanut butter and jelly or hot dogs and beans. Many cuisines have such prescriptive rules for combining their basic carbohydrate staple with a particular type of accompaniment.
In Richards' classic study of the Bemba in Africa (1939), the staple grain was considered incomplete without the relish (vegetable accompaniment). If a woman was too tired to collect the ingredients for the relish, she did not make any meal at all because she could not serve the staple alone. We will discuss several of these cuisine rules below.
Sequencing rules prohibit or prescribe the order in which foods can be eaten through time. European cuisines pay specific attention to the order of sweet and savoury foods. Sweet foods are eaten at the beginning or end of a meal or as an unstressed relish with main dish. The presentation of types of dishes can also be constrained by sequencing rules that do not allow soup to follow the main course or dessert to precede it.
Cultural rules for food use also involve the association of particular food items, prepared dishes, and types of meals with particular social events, times, places, and social contexts. The type of dish or type of menu will depend on the social occasion, the time of day, the place, and the group who are eating together.
The Importance of Studying Food-pattern Rules
Why is it important to study food-pattern rules? The major contribution of such studies is not only to describe what people eat but also to understand why. Cultural rules for patterning food intake are neither simple to list nor invariable in their operation. Like most aspects of culture, such socially shared understandings contribute to, but do not determine, choice. To identify the social mechanisms for maintaining pattern rules and to understand their range, variability, and conditional nature, it is necessary actually to study domestic and peer-group interaction.
Another reason for studying shared cultural rules influencing food intake is to understand the process of change in food systems. Often, a food system can have strongly internalized rules for assigning food items to particular individuals or events or segregating and combining foods into dishes that are restricted in use. Attempts to introduce new items into a food system are likely to fail unless there is an understanding of the rules for composing dishes out of individual foods or meals out of dishes. The acceptability of a new food may be improved if the item can be structurally integrated into a culturally familiar dish or made analogous to an old food by manipulating colour, texture, shape, or whichever sensory attribute defines the essence of the food. Understanding pattern rules can help us to comprehend how change is likely to occur and may be useful in planning interventions as well as in forseeing the unintended consequences of systemic changes in production and marketing.
Anthropological Classics
In order to study the pattern rules of the food system, a significant amount of ethnographic work, open-ended probing, and participant observation is required. However, ethnographic classics in nutritional anthropology provide no clear-cut methodological strategy to follow. Most of the early studies of food and culture were outgrowths of the traditional ethnographic method and were labour- and time-intensive. Moreover, they were often aimed at illuminating theories of culture rather than the impact of culture on intake.
As Levi-Strauss is often quoted as saying, food is not only "good to eat," it is "good to think." Food is an area of interest for theoreticians of culture because it is a material item essential for survival, and it is also heavily invested with symbolic meaning and elaborate rules for use. Food is a domain of culture that often illustrates the relationships between material factors (production and sources) and cultural symbolism involving concepts of health, relationships with the supernatural, and social relationships between the sexes and within the family, the community, and the external world.
While several anthropological classics include information about both food intake and symbolism, they focus more intensely on the latter and do not systematically tie cultural ideas about food to actual eating behaviour. For example, Firth's analysis of food as symbol among the Tikopia (1973) focuses more on the logical and systematic patterning of symbolic analogies and reversals than on actual diet. While Rappaport (1967) did collect data on intake, he was comparing the allocation of foodstuffs between pigs and humans rather than examining in detail the nutritional adequacy of human diets. His analysis of the interaction between ecological factors and the cultural definition and evaluation of food is oriented more toward explaining basic cultural processes than observing the actual operation of cultural rules in food use and individual ingestion. Similarly, Douglas's explanation of the "Abominations of Leviticus" (1966) is most concerned with cultural belief systems and less with the way in which such systems actually affect food intake. While these classics are of interest to nutritionists, they do not seem to provide enough information applicable to problem-solving policy research to justify such a time-consuming and often vague methodology.
Many early writings assumed that social systems were isolated and closed. The earliest studies actually focused on geographically isolated tribal groups in which it was relatively easy to define the available foods and look at those that were used, avoided, or preferred. These groups were studied using the developing field-work techniques of anthropology. Anthropologists cut themselves off from previous social ties and immersed themselves as participants in the '`alien" socio-cultural system. Through intensive interaction and participation in a culture in a face-to-face manner over several months the anthropologist was eventually able to describe the generalized food pattern and the folk rationalizations of the definitions of food categories and ascriptions of value. The observer was searching for the shared pattern and not the variations in the observation of rules nor the possibility of change over time.
Today, the emphasis has changed. The societies we wish to study, including our own, are undergoing massive changes in food production and distribution. There has been an expansion of the variety of available foodstuffs and channels of procurement. Increasing social and geographic mobility lead to frequent changes in the individual's socio-cultural milieu through a lifetime while the macro-system of food availability also changes rapidly.
Today, much research on food use is aimed at direct problem-solving goals related to improving the health and nutritional status of populations. It is no longer feasible for the anthropologist to spend up to two years as a participant observer in order to understand a food system. Given the flux of conditions, such continuous intensity of observation is also no longer as useful. The task for this discussion is twofold: first, to examine whatever we have learned about cultural rules for food use and, second, to illustrate how they are best studied and understood. In doing this, we must restore faith in the techniques of ethnography. Without some long-term intensive relationships with informants, without in-depth open-ended elicitation and observation, such pattern rules cannot be understood.
Most studies employ a mix of research techniques, which is common in anthropology because it has proved useful to combine the strength of intensive, qualitative data with the quantitative results of structured interviews. Qualitative data help elucidate the unique world-view and meaning system of the group being studied; quantitive data can be useful in developing scales and indices.
Food research can involve both data collected about actual (real) behaviour or the elicitation of ideal rules (the cognitive system of "shoulds" and "oughts" shared by a group). One cannot automatically infer actual food intake from stated ideal rules nor can one automatically infer an underlying system of meaning from actual behaviour. The two types of information are complementary.
Information about actual food behaviour has been collected by a variety of means in cultural food research: (a) direct observation without intrusion; (b) a combination of observation and probing elicitation for explanations; (c) dietary record-keeping for varying periods of time and with varying degrees of precision; and (d) interviews about specific events, including 24-hour recall and recollections about special events.
Direct observation without intrusion is designed to minimize the effect of the observer on the behaviour. The use of a video-tape recorder over long periods of time is one example. Another is the presence at meals of an observer who has a neutral role (Goode, Curtis, and Theophano, 1981, 1984; Goode, Theophano, and Curtis, 1984; Jerome, 1979).
Other research has involved the development of intimacy and trust with informants over a long period of time so that behaviours and their underlying patterning can be openly discussed (Laderman, 1981a). In this strategy, the unique relationship between the researcher and the informant is stressed. Such a relationship, developed over time and through mutual respect and reciprocity, is one means of breaking down social distance and learning about what really happens within households. The technique of developing such close relationships with informants assumes that, while behaviour observed early in a relationship may be guarded and artificial, the passage of time and the continuity of the relationship leads to natural behaviour at later stages, which can be especially important in regard to eating behaviour.
Direct compilation of data on people's food intakes can be an important means for identifying cultural rules. Methods range from relatively precise (and intrusive) direct observation, including weighing and measuring portions, to somewhat less precise self-reports, such as the keeping of food diaries, and structured responses, like the 24-hour-recall method.
Ideal rules for behaviour are all developed from interview techniques including, (a) casual conversations, as part of traditional long-term ethnography, (b) open-ended interviews, and (c) pre-coded survey techniques. The focus of the search for ideal rules can be on patterns, e.g. what is eaten for typical meals, on typical days, for special events, or queries about food preferences or food avoidances. Very often the most useful insight may emerge from a casual conversation in an unstructured situation rather than from a pre-structured question. Openended interviews are useful in initial, exploratory stages of research, while pre-coded surveys work best to test hypotheses derived from less structured techniques. Constructing formal measures before one is familiar with folk categories of foods and food events often yields less useful results.
Laderman (1981b), in her work with Malays, avoided a survey early in her study since she "was convinced that simply asking people how they would classify individual food items would yield nothing more fruitful than the usual mystifying contradictory charts." She wished to discover the "basic underlying criteria"' used by Malays to classify foods. Some of these were not consciously recognized and had to be drawn out by probing elicitation. Since her data dealt with item classification rather than recipes and meals (requiring observation), she was able to elicit intensive data from lengthy discussions with three key informants. She probed for "possible organizing principles" by making suggestions with which informants were quick to agree and disagree. She also probed for the reasons for her key informants' ranking or classifications of foods when she perceived them as ambiguous. In addition, Laderman picked up many "situational" comments regarding foods, since she often heard women casually discussing the appropriateness or "goodness" of foods when she was present in domestic situations. From long-term observation, she learned that time and weather influenced whether people followed shared rules or not. With an elaborate framework thus developed, Laderman was able to construct a neutrally phrased survey.
Another important facet of the strategy of cultural food research is the social unit of inquiry, which is crucial to the anthropologist. Much nutritional research focuses on asking individuals about their actual or ideal food behaviour. Two assumptions seem to be taken for granted: (a) that the individual controls his/her own food intake, and (b) that the individual is representative of a particular group based on ethnicity, class and/or locality, or status that is related to age and sex. While ultimately we ask questions of individuals, anthropologists are more likely to focus on the household as the unit of analysis, since the household is the locus of so many food decisions.
Also we are interested in determining the degree of shared communal patterning and the processes of social transmission in different sizes and types of social units. For example, it is often assumed that a macro-cuisine, such as Jewish dietary laws, exists at the social level of a major religion, tribe, or nation. One knows nothing about whether or how actual communities (localized networks of interacting households linked by kinship and friendship) hold, transmit, and reinforce these beliefs. More is known about ideal macro-cultural beliefs than their actual operation in communities. Even if there are common staples and sets of common seasonal, ritual, work, and leisure cycles in a nation or region, there are still variations that necessitate the investigation of local community differences, as Laderman illustrates for Malaysian communities (1981a,b). Knutsson and Selinus (1970) also illustrate that knowledge of a macrocultural taboo cannot lead to the assumption of universal practice or logically derived dysfunctional effects. In their study, they showed that knowledge of the periodicity and duration of Ethiopian fasting patterns as a macro-cultural phenomenon was not sufficient for understanding real nutritional implications. By collecting actual intake records in both rural and urban households, inter-community variation as well as compensating patterns were discovered.
In urban-industrial, market-dependent settings where environmental constraints are less significant and the variety of food available is great, socially mediated meaning imbued in food becomes even more important in the process of social differentiation and peer-group social transmission. Modern industrial macro-cuisine is a mass phenomenon, electronically purveyed, but we do not know to what extent these mass rules and models are used or shared by people in peer groups based on kinship, ethnicity, and class.
Researchers have also found it useful to investigate intra-community variations in food use, particularly in relation to socio-economic levels. For example, Bennett (1942) found that patterned economic and ethnic subcommunities existed in a small farming region; Jerome (1979) found subcommunity differences based on differences in economic adaptations in a group of urban immigrants with common (rural) backgrounds; DeWalt (1979) found that material and educational factors made a difference in a small Mexican village; and Goode, Theophano, and Curtis (1984) found that the degree to which households were socially embedded in the local community was related to variation in food behaviour.
When dealing with cultural food systems, it is necessary to specify clearly and differentiate between three social unit levels: the supra-local macro-cuisine, the community or peer-group pattern, and systematic subcommunity differences manifested in different types of households.
Group and individual culinary behaviours are affected by a number of extrinsic and intrinsic factors:
This paper is primarily concerned with the cultural rules pertaining to the last three factors, which are most directly influenced by the way people think about the organization of food intake. The other aspects are addressed in this volume in Messer's paper on the determinants of food intake (see chapter 1). Decisions and activities related to food acquisition involve gathering, harvesting, marketing, systems of food exchange and gift-giving, and other "pathways" by which households acquire food.
The concept of pathways of household food acquisition has been developed by Kathleen DeWalt (1979). In this aspect of the food choice process both structural factors (economic resources, time, and location of the food supply) and cultural preferences and norms interact. The cultural models that inform supermarket shopping decisions are also different from those involved in meal planning or the centrally controlled organization of food intake within a household.
Important for understanding topics of major interest in this paper, which addresses the influence of culture on the planning of meals or menu negotiation, the nature of meals, and the acts of ingestion, are long-term observations within the household context and study of peergroup influence on the focal female food-preparer. This approach is necessary for understanding not only such patterning, but also its social maintenance and its interaction with situational constraints, such as the task-allocation systems identified by Messer in her time allocation analyses (see chapter 4).