Re: Opinions as Evidence

Robin Hanson (
Wed, 26 Mar 1997 11:07:22 -0800 (PST)

Hal Finney writes:
>>it turns out to be difficult to explain persistent divergence in opinions.
>this would not establish that I should give more weight to others than
>to my own experience, because I may not be typical of average minority

But typicially you will be typical :-). Of course if you have some
unique info that others don't have, they can expect that your specific
opinion must differ from their uncertain intermediate opinion. The
hard (but not impossible) thing to explain is how they might know
you have access to unique info, and know your specific opinion, and
yet still choose to disagree.

>... this rests on my claim that most people prefer to decide
>for themselves rather than listen to other people. It could be that I
>have mostly associated with individuals who are unusual in this regard,
>or I could simply be mistaken in my perceptions. Perhaps I am focusing on
>just those areas where they differ, ignoring the vast body of accumulated
>knowledge which everyone shares. Or maybe people like to exaggerate
>their uniqueness and downplay the extent to which they follow the crowd,
>for cultural reasons. also writes:
>Persistent divergence is downright ubiquitous. The obvious conclusion
>is that most if not all people do not use even a passable
>approximation of Bayesian inference.

Is opinion divergence on matters of fact, where each side knows of the
other's opinion, really ubiquitious among ordinary people? I'm not so sure.
Consider these alternative explanations for common apparent divergence:
1. Disputes over values using the language of disputes over facts.
2. Disputes most people think are re values, but others know are re facts.
3. Incentives to signal no fear of crowd, ability to think by self
4. Forums biased toward presenting (clear n vivid) extreme views
5. Advocates with interests, eg lawyers, rewards to early idea advocates
6. A minority who like to shock & be contrary
7. A minority who consistently make the error of ignoring other's opinions
8. Random cognitive errors among pitoval people who won't revisit issues.
9. Ignorance that there are reasonable people with differing opinions

>I'd say most if not all people have their opinions driven by memetic
>replication (as exemplified by fashion and religion) and not a
>passable Bayesian approximation. ...
>Interestingly, science seems to conform to the predictions of this "community
>of Bayesian agents" model much better than society as a whole. ...
>would indicate that scientists, at least while working in their fields, are
>more "rational" than the population as a whole

Actually, the phychology of science evidence seems pretty conclusive
that the cognitive strategies of scientists are not much different
from ordinary folks.

>a) The population as a whole is often wrong on major, simple issues (e.g.

Cloning is neither simple nor major for most people, and the issues on
which opinions diverge are tied up in important ways with questions of
value, not just fact.

Robin D. Hanson