Are Beliefs Like Clothes?

Robin Hanson (hanson@hss.caltech.edu)
Sun, 13 Jul 1997 22:07:59 -0700 (PDT)


Clothes are both "functional" and social. Clothes are functional in that
they keep us warm and cool, protect us from injury, maintain privacy, help
us carry things, and enhance our figures. But since they are usually
visible to others, clothes also allow us to identify with various groups,
to demonstrate our independence and creativity, and to signal our wealth,
profession, and social status. The milder the environment, the more we
expect the social role of clothes to dominate the functional role.

Beliefs are also both functional and social. Functionally, beliefs help
us choose our actions, given our preferences. But many of our beliefs are
also social, in that others see and react to our beliefs. So beliefs can
also allow us to indentify with groups, to demonstrate our independence
and creativity, and to signal our wealth, profession, and social status.

For example, a belief that the local sports team will do well can help me
associate with my locality. A belief that my future looks bright can help
me attract a mate. A belief that UFOs are aliens could help me signal
that I'm an independent thinker, while a belief that UFOs are bunk could
help me signal my scientific education. A belief that a Democrat would
make a better president could help me signal my caring and concern about
others.

For subjects with little social monitoring and strong personal penalties
for incorrect beliefs, we expect the functional role of beliefs to
dominate. Beliefs about small military missions or small engineering
projects come to mind. But for subjects with high social interest and
little personal penalty for mistakes, we expect the social role of beliefs
to dominate. Consider beliefs about large elections or beliefs addressing
abstract philosophical, religious, or scientifc questions.

If people could wear a parka undetected underneath a bikini, bikinis might
be in fashion in the Arctic, and Arctic folks wouldn't need to trade
off the functional and social roles of clothes. Similarly, if people
could choose their external behavior independently of their internal
beliefs, then internal beliefs would never need to be anything but our
best computationally-feasible estimate of the way things really are. But
since it is typically hard to say one think and believe another, the
functional and social roles of beliefs do conflict.

O.K. Now that we realize this, what do we do? I think it is a mistake to
assume that the solution is to educate people about how to better
construct more functional beliefs. It is not obvious that, given typical
preferences over functional vs. social outcomes, people are biased toward
the social role of beliefs. Of course more education on either role can
help people get better results without hurting the other role. But more
education costs more, and it is not obvious that people are biased against
such educational spending.

The conflict between the functional and social roles of clothes has become
less of a concern with the invention of new materials and designs, and as
we have become rich enough to move to milder climates. Similarly, I think
a key to reducing the conflict between the functional and social roles of
beliefs is to better understand both roles, espcially the
academically-neglected social role of beliefs.

Just as it seems that teenage smoking can't be reduced much without giving
teenagers good substitute ways to show their independence and coolness,
the social costs of mistaken beliefs about politics and UFOs probably
can't be reduced much without giving people good substitue ways to show
their concern and independence. But it is hard to create substitutes for
things until you understand in some detail the functions they now perform.

Also, the conflict between the functional and social role of beliefs can
be muted by the invention of better social institutions, such as
political parties or academic journals, to which people might delegate
their actions. Within such institutions, social relations may be
structured so that social pressures are better aligned with functional
pressures on beliefs. But, again, it is hard to design better
institutions with only the fuzziest understanding of these pressures.

Fortunately, innovation in clothes proceeded even without a deep
understanding of either fabrics or the social and functional roles of
clothes. Similarly, belief complexes and social institutions have evolved
even without a deep understanding of the nature of computation and of the
social and functional roles of beliefs. But there is reason to hope that,
as with clothes, such evolution may proceed faster with a deeper
understanding of these issues.

Robin D. Hanson hanson@hss.caltech.edu http://hss.caltech.edu/~hanson/