2 Introduction
claim the right to deal with any manifestation of the mental or
social activity of the peoples under study that seems likely to al-
low us, in the process of the analysis, to complete or to shed
light on the myth, even though it may not constitute what musi-
cians would call an 'obbligato' accompaniment."
In the present work, a similar but reversed procedure will be
followed. Although reference will be made when necessary to
mythical and ritual data, the main emphasis will be placed on
different aspects of daily behavior that by themselves do not ap-
pear, at first, as privileged conceptual manipulations. For instance,
we shall see that Panare food is particularly good, not only good
to eat, but also good to think. Some progress in our understand-
ing of the conceptual manipulation involved in cuisine has al-
ready been made by Lévi-Strauss (1968, pp. 396-411) himself
in his now famous "culinary triangle." In displacing the emphasis
toward an implicit mythology that, in many respects, is more
acted out than spoken in the flow of daily behavior, we shall be
able to explain a number of facts that could not be explained
otherwise and remain generally as incongruous residuals in the
notebooks of the anthropologist. Being implicit, this mythology
evidently cannot be immediately perceived by the observer. This
is because the symbols that it manipulates are not given per se,
as in myths, but are always conglomerated with other facts, the
rationality of which is to be found elsewhere: in the ecology, in
the level of technoeconomic development, that is, in the infra-
structure (in the Marxian sense) of the society under consider- .
ation. Ultimately, what we shall look for is the symbolic element
that is to be found in each fact, since it is that which will allow
us to discover in turn the conceptual system through which the
Panare conceive of themselves.
To reach this objective, we shall rely upon the structuralist
method, as spelled out in Lévi-Strauss's Structural Anthropology
(1968). Since the literature on structuralism has become avail-
able in English in the recent past (see, among others, Lane 1970,
Ehrmann 1970, Hayes and Hayes 1970, Leach 1970, and, above
all, Piaget 1970), I need not explain the method but need only
summarize some of its principles.
A structure is a supraempirical construct that therefore cannot
be observed. What is observed in the field are facts or events that
are themselves the products of a structure. The facts can be con-
sidered as messages that are expressed according to a certain code.
The analysis consists therefore in breaking the code, in decoding