Wednesday, January 08, 2014

Linguistically challenged by the expression "believe in"

The issue is confounding two distinct domains - one where a held view is a matter of faith, the other where empiricism rules and theory is evaluated by its consistency with the evidence.  Concern about this confounding was expressed yesterday in a post called Distinguishing Science From Nonsense, which appeared in The Chronicle's blog series, The Conversation.  Similarly, over the last week or so I've seen much dismay expressed over this recent Pew survey regarding the American Public's View on Human Evolution, where it was found that about one third of the populace are of the mind that human beings have existed in their present form since the earth began.

I will return to human evolution later in this post.  I want to begin by taking a look at the first domain and why people who consider themselves scientifically literate nonetheless indulge in matters of faith, particularly of a non-religious nature.  So, for example, parents of young children will encourage their kids to believe that Santa Claus will come down the chimney on Christmas eve to stuff their stockings and that the tooth fairy will put money under their pillow in exchange for the lost tooth.  When I was ten, the Lovin' Spoonful came out with a hit, Do You Believe In Magic.  A couple of years later on the Sgt. Peppers album, the Beatles at their best in my view, there is this wonderful song, With A Little Help From My Friends.   Within it are the lines:

Do you believe in love at first sight?
Yes I'm certain that it happens all the time.

And to round out this set of examples, consider that many sports fans, myself included, subscribe to The Jinx and the particular manifestation known as the announcer's curse.

I deliberately chose this set of examples because most people would find them benign, if not delightful.  They conjure up a romantic view of the world.  Most people want to cordon off parts of their lives to hold romantic views within that universe.  On the language issue, my preference would be to restrict the expression "believe in" to romantic or religious uses and to use something else entirely for empirical matters.

The problem is that on empirical matters one can hold an incorrect theory to be true, e.g, the ancients held that the earth was the center of the universe.  Though Keynes may not have actually said this line, he is frequently attributed with having said:

When the facts change, I change my mind. What do you do, sir?

This brings us to the heart of the matter.  When a person confronts evidence that brings a held view into question, must the person reconsider that view in light of the evidence?  Or is the person entitled to disregard the evidence because the person "believes in" the espoused theory?

At this point in the argument it is useful to segue either to a discussion of information literacy, most of us (and I count myself here) don't follow best practice, or to a discussion of Daniel Kahneman's book, Thinking, Fast and Slow.  Let me do the latter, very briefly.  Human beings are inherently lousy statisticians.  The fight or flight instinct can be triggered by a false positive.  More generally, the species is subject to a variety of biases that enable quick conclusions from scanning data.  Often these inferences are incorrect, but mainly we are unaware of our errors. 

One particular bias, relevant to the discussion of evolution, is that we have a need to impute causality after the fact, when before the fact we couldn't have predicted the outcome at all and hence randomness provides a better explanation.  It is not just unsophisticated people who do this.  In his book Kahneman takes Jim Collins to task for just this sort of cherry picking after the fact.  Collins is the author of Good to Great, a very highly regarded book about business success.  I am now reading What the Best College Students Do by Ken Bain.  In my view, Bain also engages in this sort of after the fact cherry picking.  It is a professional hazard.

Let's keep that in mind and turn to the typical citizen's view of evolution and with that let's focus on the two thirds of the population who ascribe to an evolutionary view of human beings in some way.  Abstracting entirely from the issue of God's role or not, my guess is that those who say they believe in evolution nonetheless have an incorrect view of matters.  To illustrate why, consider the picture below.  Something of this sort is often used to depict the evolution of human beings, who have "descended from the apes." 

man descending from the apes

When read from right to left the picture may be an acceptable graphical shorthand for the historical record - the evidence that points to an evolutionary view of human beings.  But read in the opposite direction, the same direction that we read text and hence the more likely way to read the picture, the image points to a linear causal view of evolution.  That view is wrong, though it is what most people probably have come to understand evolution to mean.

A quite readable book on the topic is Stephen Jay Gould's Bully for Brontosaurus, a collection of essays by Gould and by other authors writing on related issues.  Gould takes issue with how evolution is depicted in most high school textbooks - as a tree trunk.  Gould would prefer a different metaphor - a bush with many branches.  The idea is that some branch becomes favored, by chance rather than because it is inherently better a priori.  The less favored branches begin to die out.  The favored one grows and then has more branches, after which the process repeats.  This is how natural selection works.

A related mistake that people make is to equate natural selection, a neutral term, with the expression, survival of the fittest, that conveys the sense of getting stronger over time.  It is a misnomer, at best.  One of the essays included in Bully for Brontosaurus is Paul David on Qwerty.  It is included in the collection because the economics of increasing returns has quite similar dynamics to that of evolutionary biology.  The qwerty keyboard was designed to slow typists down so the manual keys would not get stuck.  A survival of the fittest view would suggest that when manual typewriters improved enough in their mechanisms, or when electric typewriters came into existence, or word processors, or personal computers, that the keyboard would change to allow more rapid input.  But it didn't.  We are locked into qwerty.  Likewise we are locked into an academic calendar that was originally set so students could go home for the spring planting and return to campus after the fall harvest.  This in spite of the fact that students don't do these things any more.  These type of outcomes can't be produced from a linear causal model. 

How can citizens have scientific literacy when they are such poor interpreters of data?  I don't know.  I fear that scientific literacy has come to mean neither familiarity with the data that scientists confront nor with the theory scientists espouse, but rather only with the conclusions scientists report, and then only in highly abbreviated form.  Given the increased specialization in the world in which we live, some of this may be necessary.  But in that regard I fear the continued use of the expression "believe in" contributes to illiteracy, by turning people into fans (as in sports fans) when they could instead achieve some basic understanding.  It's the latter we should be after.  Measurements such as the Pew survey don't do enough to distinguish between the two. 

No comments: