The excellent (and mindbogglingly productive) Megan McArdle has a piece in The Atlantic Online about bad numbers which set me thinking again about the role of blogging. OK, all by herself Megan McArdle is a cogent argument for blogging. But what she says implies a place for anyone at all on the Web who is interested in putting out reliable and trustworthy information.
Her thesis about numbers is that they stick in the mind and distort our judgment even when we know better, even when they’re wrong or irrelevant. Despite what ought to be disabling criticisms of its methodology, the first widely-publicised estimate of the number of civilian casualties caused by the Iraq War remains influential. Why? It was first in the field and came with the imprimatur of The Lancet. So the World Health Organisation’s much lower estimate – one quarter of the number – trails in the shadows.
Ideology figures here: the anti-war Left loved the Lancet figure, and no-one committed to an all-embracing ideology ever has much regard for plain old empirical truth. But McArdle cites work in cognitive psychology that shows that we’re all prone to this error. It’s a form of the well-known and well-attested primary availability error: the first thing we see of a particular kind colours our view of that kind. I like to illustrate that one by reminding people how kids cathect proper names – ‘Ooooh yuk, Derek.’
The Web is full of junk so it’s very tempting for reflective people to dismiss it as a venue for discussion. What the phenomenon of bad numbers indicates, however, is the need for good ones.
Does the same thing apply to arguments themselves? To opinions? Anecdotes-with-tendency? I hope so. What makes me almost sure is the work on attitude formation. This shows that they are rarely formed by one event. (We do get over the fact that Derek stole our peanut butter sandwich in grade two.) Attitudes take shape gradually, solidifying out of a cloud of particles of information, events and experiences small and dramatic.
So even the tiniest blog can add a stone to the pile. Well, a pebble anyway.