Culture as knowledge

If culture is learned, then much of it can be thought of in terms of knowledge of the world. This does not only mean that members of a culture must know certain rules or be able to recognize objects, places, and people. It also means that they must be able to make inferences, understand the world, making inferences and predictions. In a famous statement summing up what we might call the cognitive view of culture, Ward Goodenough wrote:

> “…a society’s culture consists of whatever it is one has to know or believe in order to operate in a manner acceptable to its members, and do so in any role that they accept for any one of themselves. Culture, being what people have to learn as distinct from their biological heritage, must consist of the end product of learning: knowledge, in a most general, if relative, sense of the term. By this definition, we should note that culture is not a material phenomenon; it does not consist of things, people, behavior, or emotions. It is rather an organization of these things. It is the forms of things that people have in mind, their models for perceiving, relating, and otherwise interpreting them.”

> (Goodenough [1957]1964: 36)

There is a linguistic homology at work here. To know a culture is like knowing a language. They are both mental realities. Furthermore, to describe a culture is like describing a language. Hence, the goal of ethnographic descriptions is the writing of “cultural grammars” (see Keesing 1972: 302 and section 6.3.2).

3 See Lehrer (1974) for a discussion of the theory of semantic fields in lexical analysis. Tyler (1978) contains detailed discussions of different models of lexical analysis within linguistics.

---

## Theories of culture

In the cognitive view of culture, the body of knowledge necessary for competent participation in a community includes both propositional knowledge and procedural knowledge.

**Propositional knowledge** refers to beliefs that can be represented by propositions such as *cats and dogs are pets, smoking is bad for your health,* and *newborn babies cannot crawl.* These are the “know-that” types of statements ethnographers often try to elicit from informants. **Procedural knowledge** is the “know-how” type of information that must often be inferred from observing how people carry on their daily tasks and engage in problem-solving. To drive a car we not only need to know what different parts of the cars do, e.g. a certain pedal if pressed increases the speed or stops the car (propositional knowledge); we also need to actually know when and how to use that information. We need to know the “procedures,” that is, the specific sequences of acts, through which a given goal, for instance, accelerating or stopping, can be achieved. We also need to recognize whether a situation requires a certain action.

In the 1960s cognitive anthropologists became interested in terminological systems as a way of tapping into the cognitive world of a given group of people:

> “To the extent that cognitive coding tends to be linguistic and tends to be efficient, the study of the referential use of standard, readily elicitable linguistic responses – or terms – should provide a fruitful beginning point for mapping a cognitive system. And with verbal behavior we know how to begin.” (Frake [1962]1969: 30)

Language in this case is understood as a set of propositions about what the speaker (as a member of a society/speech community) knows (or believes). Such propositions must all be reduced to the form: Subject + Predicate, e.g. *this plant* (Subject) *is a strawberry bush* (Predicate), *John* (Subject) *is Mary’s father’s brother* (Predicate), *a hibiscus* (Subject) *is a kind of flower* (Predicate). Such propositions can then be connected to larger sets through rules of inference like the following:

John is Mary’s father’s brother

x’s father’s brother is x’s uncle

John is Mary’s uncle

Cognitive anthropologists rely then on the knowledge of linguistic categories and their relationships to show that to be part of a culture means (minimally) to share the propositional knowledge and the rules of inference necessary to understand whether certain propositions are true (given certain premises). To this propositional knowledge, one might add the procedural knowledge to carry out tasks such as cooking, weaving, farming, fishing, giving a formal speech, answering the phone, asking for a favor, writing a letter for a job application.

In more recent work on culture and cognition, the task of finding cultural “rules” on the model of linguistic rules has been abandoned in favor of models that are said to be less dependent on linguistic formalism and linguistic analysis (Boyer 1993a; Dougherty 1985). Psychologists, philosophers, and anthropologists have argued that there are categorical schema (or schemata) that are readily available to the human mind and these form natural kinds, categories about which people seem to be able to make inferences without having an explicit “theory” or “model” of such concepts. The approach earlier advocated by ethnosemanticists like Frake or Goodenough does not seem to work because people are not able to provide the propositions (or the features) that describe the necessary and sufficient conditions for what constitutes a “dog” or a “shaman,” but they can consistently show that they have an intuitive understanding of what these concepts imply. Even young children can easily infer that something that was referred to as a dog eats food, sleeps, and looks at things, whereas an object like a hammer cannot engage in any of those activities. One of the most commonly mentioned example of a natural kind is “living kind” (Atran 1987, 1990; Atran and Sperber 1991; Sperber 1985). The fact that children seem to easily acquire an understanding of living-kind terms without being taught and with very little direct experience has been used as evidence that there are “innate expectations about the organization of the everyday biological world.” (Atran 1993: 60) According to Atran, one of these expectations is that living kinds have an essence whereas artifacts are defined by functions.

This theory about innate ability to make categorial distinctions has been variously used by symbolic anthropologists interested in ritual and religious life (Boyer 1990; Boyer 1993b). Bloch (1993), for example, utilizes Atran’s hypothesis about the naturalness of the living-kind category for a rather complex argument about how the Zafimaniry of Madagascar can conceptualize the transformation of human beings into artifacts (the houses they used to inhabit). After the death of the couple who built it, a house is seen as the couple and becomes a “holy house” (*trano masina*), a source of blessing for the descendants (Bloch 1993: 115). To understand this symbolic transformation, Bloch argues, we must take into consideration the fact that before becoming “wood,” the material with which the house was built was trees. “This passage from people to trees was possible in the mind because it is premised on the unity of the domain of living kinds” (Bloch 1993: 119). The further passage, from living kind (trees) to artifact (house), is however more problematic or less natural for the human mind and therefore, Bloch argues, needs material symbols, including massive decorated wooden planks which replace, over time, the flimsy parts of the house (woven bamboos and mats) used by the original couple. The central posts and hearth become then the permanent replacement of the ancestors and it is these artifacts that the descendants address when seeking a blessing.⁴

Although this new generation of cognitive anthropologists claim to be less dependent on linguistic analysis than their predecessors, the shift of focus from the description of separate cultural systems to the universal bases of human cultures reproduces the shift from behaviorist to innativist theories of language in the last thirty years. Chomsky (1965, 1968) argued for innate principles for language acquisition based on the fact that children do not have sufficient input to be able to produce the type of generalizations they need to acquire the fundamentals of language in a relatively short time (two to three years).

 Similarly, contemporary cognitive anthropologists argue that for certain types of cultural concepts, there is not sufficient evidence in people’s experience. For example, religious symbolism tends to involve implicit principles — principles that are rarely fully articulated — and vague statements. Hence their acquisition would not be possible “without having certain principles that make it possible to go further than the material given” (Boyer 1993: 139). Such principles consist of the application of assumptions about natural kinds to a non-natural domain. According to Boyer, much of religious practice is made possible by the construction of such “pseudo-natural kinds.” This simply means that many cultural categories (e.g. what constitutes a shaman, a poet, or anyone who has some special, undefinable characteristic) are used “either directly as natural-kind names, or as a predicate which implies the existence of a natural kind” (Boyer 1993: 132).