Part of the reason I am interested in decision theory and puzzles is that so many decisions in life are made irrationally and sub-optimally, while many puzzles go unsolved. Anyone who spends any time considering the ease with which people can be confused by numbers and faulty reasoning risks despairing for the future of the human race. The saving grace is that we made it this far, and occasionally we hear a voice in the wilderness calling out the truth.
In this case, the truth is in a book by Charles Seife, Proofiness. You might know him from a previous book, Zero. Both are must-read volumes. The word “proofiness” is a made-up word to describe a way of supposedly proving a point, but not really doing it. Proofiness is not always a swindle; sometimes the perpetrator simply does not know better, but we often see examples in politics and religion (where else?). Seife gives many examples including ones that tear the august Supreme court to shreds by pointing out the obvious proofiness of decisions that trace back to politics. [BTW, “politics” is derived from the Greek poly for many, and a tic is a blood-sucking parasite–anon]
Seife starts off with a typical example of a very simple scam. The junior senator from Wisconsin, McCarthy, held up a sheaf of paper and announced he had the names of 205 known communists who were working in the State Department. If he had simply said something like, “I know many communist sympathizers have infiltrated the government,” he would have likely been ignored, but the specific charges of a specific number of agents in a specific organization gave him enough credibility to start a House investigation that is infamous to this day. The fact is he had no list. He had no numbers. He had no names. When the publicity took off, he had to scramble and enlist the help of conservative newspaper owners to research and get him some names. Even with this help, he never did get a list of anywhere near 205, and the list he got was bogus.
Variations of this scam are all around us. “Nine of out of ten doctors prefer ________ [fill in the blank].” This is a double scam because we do not know how many doctors were surveyed or what they prefer _______ to. This is, the sentence is incomplete and you are expected to fill in the blank with “to everything else.”
Simply speaking forcefully with definite data and numbers will often win an argument even if you make the data and number up out of nothing. This can be fun with friends, but not so fun in a courtroom. And do not think that judges are more logical than common folk, Seife gives examples of judges using scary twisted logic.
I will probably post articles on several aspects of “Proofiness” starting with one of my favorites, the fallacy of fair voting. Readers of this series and my other blogs from years back know this is one of my hot buttons. Folks, voting is a physical measurement! A census is a physical measurement. They should both be treated as such. Did Bush or Gore win Florida? No. Neither won. It was a tie. Any other solution has nothing to do with physical measurement and a lot to do with politics. That is the way the game is played, but let us not pretend to be sanctimonious about it. Whoever is behind in a recount will argue against dis-enfranchising voters. Whoever is leading, no matter how slim the margin, will argue against allowing the opponent to steal the election by allowing obviously invalid ballots. Neither will come out and say, “It is a tie, but I really want to win, and I will use any trick I can get away with to ensure that.”