malaria in Africa

[Gautam Mukunda]

Charlie Bell wrote:
Heh. It's very frustrating to people who don't have spare time, and  
it's very frustrating to people who are trying to sort out one point  
to be totally smothered. You're not "exploring details", you're just  
drowning people in volume, and switching or adding topics. It's very  
poor in debate, and it's just plain rude in a conversation. But after  
10 years, I'm pretty sure you're not going to change.

I replied:
Look, Charlie, Dan is fantastically good at researching and analyzing data.  There's something frankly perverse in the idea that such an ability (one that puts him in the tiny handful of the very best I've ever met at such things) is something that he should _not_ use on the list.  He's not "smothering" you with data, he's doing data analysis.  There are basically two ways to construct a logical argument.  You can be inductive (reasoning from concrete details into general findings) or deductive (reasoning from general theories into concrete hypotheses).  Dan is very good at both, but when he's reasoning from evidence he's engaging in superb inductive reasoning.  Quite often it's good enough that it's basically a model of how to construct an argument, one I would use enthusiastically if I were teaching a class on the subject.  If he's not allowed to use data to support an argument, exactly how is he supposed to try to persuade someone?  I find inductive
 reasoning in politics to usually be vastly superior to deductive reasoning, because it is empirical and because our theories of politics are insufficiently well-grounded to value them over countervening information.  Empiricism requires data.  If you're not as good at it as he is (no shame - I'm not either) I would think reading and debating with him would be a great opportunity to _get better at it_.  If he challenges your opinions using data it might be worthwhile once in a while to consider whether your opinions should change, instead of believing that he has bad motives.  What you call changing topics is usually, for example, use of an enormously valuable technique - drawing out the logical implications of stated beliefs into a different domain and seeing if they still make sense.  If they don't, they probably don't make sense in the first domain _either_.  How do you try to persuade people to change their minds?  And in particular, how do you do it
 without using data?  For example, in this discussion I have _not once_ seen anyone actually engage with the argument or the data.  There are dismissals any point of view differing from the priors as bought and paid for (I've always wanted to ask people who believe that - if you think everyone's opinion is for sale, doesn't that really say something about yours?).  I've seen cites to irrelevant arguments (DDT is nasty - well, no shit.  It's an insecticide.  Is it as nasty as malaria?  Is it as nasty as the chemicals that might be used instead of it?).  And I've seen no concern whatsoever with the people involved - like his daughter.  Dan is a real scientist, and I'm at least a social scientist, so we're both trained to ask a simple question in any argument  - what is the obtainable information that would cause you change your belief?  If you can't come up with an answer, haven't you just said that you're not persuadable at all?  And if you _can_, why do
 you reject as ill-intentioned (and what would his motives be, exactly, for having ill-intent?) efforts by a very bright and talented person to bring such information to bear?

Gautam


      ____________________________________________________________________________________
Looking for last minute shopping deals?  
Find them fast with Yahoo! Search.  http://tools.search.yahoo.com/newsearch/category.php?category=shopping