THE TRIANGLE
SI7
It
would seem that the methodological principle which inspires
such distinctions is transposable to other domains, notably that of cooking
which, it has never been sufficiently emphasized, is with language a
truly universal form of human activity: if there is no society without a
language, nor is there any which does not cook in some manner at least
some of its food.
We will start from the hypothesis that this activity supposes a system
which is located-according to very different modalities in function of
the particular cultures one wants to consider-within a triangular seman–
tic field whose three points correspond respectively to the categories of
the raw, the cooked and the rotted. It is clear that in respect to cooking
the raw constitutes the unmarked pole, while the other two poles are
strongly marked, but in different directions: indeed, the cooked is a
cultural transformation of the raw, whereas the rotted is a natural trans–
formation. Underlying our original triangle, there is hence a double op–
position between
elaborated/unelaborated
on the one hand, and
culture/
nature
on the other.
No doubt these notions constitute empty forms: they teach us noth–
ing about the cooking of any specific society, since only observation can
tell us what each one means by "raw," "cooked" and "rotted," and we
can suppose that it will not be the same for all. Italian cuisine has
recently taught us to eat
crudites
rawer than any in traditional French
cooking, thereby determining an enlargement of the category of the
raw. And we know from some incidents that followed the Allied land–
ings in 1944 that American soldiers conceived the category of the rotted
in more extended fashion than we, since the odor given off by Norman
cheese dairies seemed to them the smell of corpses, and occasionally
prompted them to destroy the dairies.
Consequently, the culinary triangle delimits a semantic field, but
from the outside. This is moreover true of the linguistic triangles as well,
since there are no phonemes
a,
i,
u
(or
k,
p,
t )
in general, and these
ideal positions must be occupied, in each language, by the particular
phonemes whose distinctive natures are closest to those for which we first
gave a symbolic representation: thus we have a sort of concrete triangle
inscribed within the abstract triangle.
In
any cuisine, nothing is simply
cooked, but must be cooked in one fashion or another. Nor is there any
condition of pure rawness: only certain foods can really be eaten raw,
and then only if they have been selected, washed, pared or cut, or even
seasoned. Rotting, too, is only allowed to take place in certain specific
ways, either spontaneous or controlled.
Let us now consider, for those cuisines whose categories are relatively
well-known, the different modes of cooking. There are certainly two
