“Did you know that 67% of all statistics are made up on the spot?”
Some would see this as the joke that it is. To others, though, it’s a distortion and a complete abuse of mathematics. You probably see the joke. Mr. Seife would not get it.
Seife’s new book, Proofiness, takes any mention of numbers as very serious things. He argues that we are doomed because people are so easily gulled into believing statements that contain a number. For a proof, he cites Senator Joe McCarthy’s use of the number “205” in McCarthy’s false charges of communists in the government. Seife thinks people believed Senator McCarthy because McCarthy used a number, and not a noncommittal phrase.
He further illustrates his point that numbers are serious with the museum guide who claims that a skeleton is 65,000,038 years old. To wit:
“How could you possibly know that?” the teenager shoots back.
“It was 65,000,000-years old when I started and that was 38-years ago.”
Cute. But so what?
It is these examples, expressed light heartily, that Proofiness uses to tell us that Americans are so gullible as to believe anything with a number attached. If you agree with that, the rest of this book is simple commentary.
Seife does use some serious examples to show where mathematics can be abused but again, I would ask, so what? He shows a graph to demonstrate that counties that use more power have citizens who live longer. This is, as he notes, a case of correlation, not causation. More power consumption will not make a person live longer. But living in a developed country that uses power is a dead giveaway for a longer life.
Seife turns to serious business when he discusses elections but his arguments are flawed. On elections, he seems to confuse mathematics with laws passed by a legislative body. Mathematics is precise: two is a number and is different from e. However, in elections, the mathematics is not generally at issue, although people like to couch it that way. Elections are based on legal definitions, not mathematical ones. The shape of a voting district, for example, is rendered by law, not by mathematical logic. Gerrymandering is not a mathematical issue, it’s a political one. (I must admit that some gerrymandered districts do resemble fractals.)
Thus when the book discusses how better to run an election based on science, the author misses the point. The problem with elections is not the science, it is that the legal requirements are often difficult for non-lawyers to comprehend. But, once the legal setting is determined, the votes cast, the elections become not a matter of scientific debate, but of legal decisions. As an (external) example of legal determination compared to science, take note of tomatoes. Tomatoes are scientifically a fruit, but for cooking they could be considered a vegetable (see http://www.oxforddictionaries.com/page/tomatofruitveg). For some matters of taxation — a legal issue — tomatoes are considered a vegetable. Scientific? Of course not. Legal? Well, sure.
The book does have some good points, for example when Seife shows the mess that non-mathematicians can make with equations. Here are a few the book cites:
Happiness = P + (5×E) + (3×H)
Misery = 1/8W + (D–d)3/8 × TQM × NA
Calipygianness = (S+C) × (B+F)/(T–V)
(If you’re wondering, Callipygian means “having well shaped buttocks.”) The equations are clearly meaningless. Who can measure emotions like “happiness” or “misery” definitively? But Seife’s point is a good one on this score. The equations do seem to add weight to discussions and the less mathematically inclined person could be persuaded that there’s science lurking in what is really a vacuous formula. I wish there had been more of this in the book.
The overall thrust of the book is that mathematics and science can be misused to present the unimportant as important; the mundane as relevant; the fictitious as factual. These are not new stories, they are well known. Proofiness gives us very little that’s new and its argument, while valid, is old and well-known to any scientist.
Would a non-scientist then benefit from this book? I doubt it. It would be better for a such a person to read a statistics text. After all, statistics never lie.
David S. Mazel received his Ph. D. from Georgia Tech in electrical engineering and is a practicing engineer in Washington, DC. His research interests are in the dynamics of billiards, signal processing, and cellular automata.