and others are not. We can usually work on the basis of such "gut feel-
ings" because they are the result of subtle calculations in our special-
ized systems, although these cannot be accessed for conscious inspec-
tion.
Now, in some contexts, the social relations people build on the
basis of these coalitional intuitions are made much easier by the fact
that the groups are defined as essential. In the laboratory studies, peo-
ple were given an arbitrary coalition to cooperate with, and as a result
they started imagining essential differences between groups. But in
real social life the opposite is very often the case. People are presented
with social categories that seem essential—castes of blacksmiths or lin- [289]
eages—and use them for coalitional purposes.
Because they are extremely stable coalitions, these essence-based
groups do not seemto be coalitions at all. That is, for all the members
as well as outsiders, the alleged essence is what drives people's behav-
ior. But I would suspect that actual behavior is more directly driven by
people's coalitional intuitions. Take for instance the case of lineages in
an African group like the Fang. Lineages are certainly defined in
essentialist terms not just in the sense that people are putative descen-
dants of a single ancestor but also because they are said to share some
essential features. When I worked in Cameroon, people frequently
told me that the so-and-so's were easygoing, the such-and-such's were
troublemakers, etc. But Fang lineages are also a place of intense,
unconditional cooperation, and you can rely on comembers of the line
to help you out, although you should not expect similar solidarity
from nonmembers, even though they may live in your village. Now
Fang lineages span territories so huge that everybody has lineage
"cousins" they seldom interact with. In these rare cases, essentialist
understandings of lineage would suggest that you can trust them any-
way (these people are the same substance as you are, you know their
personality type and therefore their reactions) whereas coalitional
intuitions would recommend caution (since this is a first-time interac-
tion and will probably remain a one-time event, why should they do
you any favors?). People in such cases generally follow their coali-
tional intuitions but then reconcile this with their essentialist concepts
by saying that they are not in fact certain that these people really
belong to their lineage.
So the point is that some social categories appear to be based on
essences—essential concepts are the ones that come to mind when think-
ing about them—but actual behavior is piloted by more complex
WHYDOCTRINES, EXCLUSION AND VIOLENCE?