Anwyn Enters the No Tarradiddle Zone

How should a state certify its petitions? Oregon's civil-union law just got smacked by their Supreme Court a Federal district judge, and whatever you think of the merits of the law, the question of whether the state followed its own election laws respected voters' fundamental rights is a valid one. Is a statistical verification of a small portion of a petition's signers sufficient? There's an argument to say it is--modern politics has gotten so unwieldy and deals with such great numbers of people that it's a tempting shortcut.

Anwyn thinks otherwise, as did the court, and I agree. It's not a poll, and we're not measuring statistical voters--we're measuring real voters and their expressed preference at a particular time. It's not just in this issue but in others--I recall a perennial debate about the U.S. Census, with all its political implications, and whether it should be compiled by statistical estimates based on population density and the like, or by counting real, actual heads. There's nothing wrong with a statistical survey of an area, particularly to estimate the effectiveness of door-to-door census taking, but again I think it's crucial that legally and politically significant decisions be based on actual counts and actual signatures of actual people, rather than on estimated totals, no matter how elegant the math underlying them.

Especially in things like petitions and elections, where one vote can make the difference and a statistically insignificant margin is politically extremely significant.

Bonus points to Anwyn for her use of the term "tarradiddle", which I thought was some sort of euphemism for nookie in Gone With The Wind.

CORRECTION: It's actually a federal court calling for a timeout, not the state supreme court as I had written, which means I was wrong only statistically correct (and my explanation of the legal issue was, um, wrong well within the blogger margin of error). Relevant discussion here.