Saturday, December 31, 2016

Prevaricators' Paradise, or Victims' Vengeance?

kw: book reviews, nonfiction, logic, lying, critical thinking

I'm sure you've heard the old chestnut about public figures who lie, that some lie when they must, and some lie constantly to stay in practice. Or that for some, that the way to tell they are lying is that their lips are moving. And then we have Mark Twain remarking on "Lies, damned lies, and statistics", which he attributed to Disraeli, but he was mistaken (or, perhaps, lying!). And we occasionally hear a paraphrase of "To err is human" into "To lie is human". Ah, the great benefits we seem to attain by lying, at such a little cost!

Is it safe to say that nearly anyone trying to advance a viewpoint (or sell a product or win an argument) is lying? Usually, I suppose. It is your best, first assumption. But when you want to know "the truth", where do you turn? Now that the biggest news story closing out the year 2016 is "Fake News", where do you get "Real News"?

We have to take what we can get, and verify it the best we can. This is the primary reason for learning to think critically. It is why right, right now is a great time for a book to appear with the title A Field Guide to Lies: Critical Thinking in the Information Age. To quote the opening paragraph of author Daniel J. Levitin,
"This is a book about how to spot problems with the facts you encounter, problems that may lead you to draw the wrong conclusion. Sometimes the people giving you the facts are hoping you'll draw the wrong conclusion; sometimes they don't know the difference themselves. Today, information is available nearly instantaneously, but it is becoming increasingly hard to tell what's true and what's not, to sift through the various claims we hear and to recognize when they contain misinformation, pseudo-facts, distortions and outright lies."
Dr. Levitin jumps right over "lies and damned lies" to statistics in the first third of the book, "Evaluating Numbers". It is a great place to start because we are so damnably bad at evaluating numbers. The brain is an unbeatable pattern recognition machine. It excels at finding similarities, and it is over-tuned: it finds patterns where none exist. Thus we see all kinds of animals and scenery in clouds; a highly educated astronomer such as Percival Lowell spent much of his career mapping the canals on Mars; and to my recollection there are at least three pieces of burnt toast being carefully stored in freezers because they bear a mark that looks sort of like Jesus, so they have become objects of worship. But we are woefully deficient in discerning the meaning of numerical quantities.

It is curious: Recognizing a face reliably is something we do thousands of times daily, and we are very, very seldom mistaken. We almost never mistake a not-face for a face. What we do easily our computing machines do slowly and rather poorly. I have made heavy use of the Picasa face-tagging system for several years, for organizing my photos. It is really, really good, but will on occasion put a hopeful "Who is this?" tag on what I see as some random blob, or on a hubcap or the "face" of a clock. And it misses about a tenth of the genuine faces in an average scene, particularly when there is quite a range of sizes, that is, distances to the subjects. The computing power needed for Picasa's algorithms to work has only been available for about ten years. It takes a few billion computations to locate a face in a scene, and more billions to match it with a similar one in the database of recognized faces in your photo collection. Every one of us can recognize a person with great reliability in a tiny fraction of the time. But very few of us can determine the product 347x94 in less than several seconds, and probably not at all without the help of a calculator or a pencil and paper. Yet a computer can perform millions or billions of similar calculations every second, and does so whenever you recalculate an Excel spreadsheet. We must face the fact that our brain is tuned to see the tiger in the grass so we don't get eaten. Prior to the ascent of "civilization", we had no need to count the tiger's stripes, and if we were interested in doing so (unlikely!), it was only after we'd ganged up on the tiger to kill and skin it.

Part Two is "Evaluating Words", and in just under 60 pages, is the briefest of treatments of logical fallacies. There are numerous volumes written about the few dozen fallacies of formal logic that philosophers study, and further volumes that cover the fallacies of informal logic, the "everyday fallacies" such as "50,000 dentists agree" or "All the Kardashians (or celebrity of your choice) love product X" or "only a fool would pass up this offer". Pet Advertising Peeve of the year: any ad containing the phrase "don't want you to know" is fallacious. Period. Trust your uncle Polymath (but, then, this is possibly another fallacy!).

The third part of the book, "Evaluating the World", presents an excellent review of the Scientific Method, which is an armory containing our best weapons to discover error and falsity and correct them. The scientific method is collaborative; it cannot be practiced in isolation. The great difficulties we encounter using the scientific method glaringly illustrate just how prone we are to being deceived and to deceiving ourselves. Thus, if the brain is a pattern-recognition engine, the scientific method harnesses many brains to sharpen the focus and more properly set the boundary between "perceived facts" and "true facts".

I hesitate to go further. A book like this presents a quandary. While it exposes a few hundred ways you could be lied to, it is also a kind of prevaricator's textbook. I am reminded of the dilemma faced by missionaries to Hawaii, as described in James A. Michener's enormous novel Hawaii: They wanted to put up a sign with the Ten Commandments, but there was a problem. The Hawaiians didn't understand them, particularly "Thou shalt not commit adultery?" What was adultery? The missionaries argued thus: If we try to describe adultery in general terms, some people will say, "Oh, that isn't what I am doing." But if we list all the different things that are adultery, others will read the list and say about one item or another, "Wow, that sounds like fun. I'll try it!" They were further troubled by the cultural requirement that the king of Hawaii must marry his sister. It took the missionaries about two generations to change Hawaiian culture enough to eliminate "that kind of adultery". Thus there is a risk that some will read the book to get ideas.

I suppose Dr. Levitin has thought the matter through, and judged that putting weapons into our hands is worth the risk that a few liars might become better liars as a result. I agree. Caveat emptor is still the best policy, and always will be. Let this book full of caveats arm you.

No comments: