The Feminist Bank Teller

Reading "Thinking Fast and Slow" has been fun, but frustrating. Of course the author,  Daniel Kahneman, deals in areas we might now call "psychology" or even "philosophy" but most often this is called a problem of "logic." Consider the problem he and his colleague, Amos Tversky, gave test subjects:

Linda is 31 years old, single, outspoken, and very bright. She majored in philosophy. As a student, she was deeply concerned with issues of discrimination and social justice, and also participated in anti-nuclear demonstrations.

Which is more probable?

1. Linda is a bank teller?

2.Linda is a bank teller and is active in the feminist movement.

If you answer #1, you get the question right and go to Harvard. If you answer #2, you get the question wrong and wind up selling tires at "Town Fair Tire" the rest of your life.

This is because you got fooled by all the information they gave you which developed a picture of the woman in your mind, but the question was not about what you had pieced together about her, but it was about probability.  If you had broken the question down into "which is more likely, that one thing is true about a person or that two things are true about that person?" well then everyone would have said, well one thing is more likely than two.

But where's the fun in that?

In fact, in literature and all sorts of performing arts, one small detail about a person is used to stand for a more complete picture of the person. Oh, she's the type who wears love beads and a flower in her hair, earth mother. From that you can expect a whole raft of things about her, that she is probably going to be tolerant of gay marriage, may bake her own bread, drive a hybrid car, wear T shirts with  peace symbols printed on them.

We all understand there are risks in making these assumptions, but we use them as a short cut to understanding more about a person than we have time to elaborate. 

But the logical problem here is called "conjunction fallacy" in which the mistake is to assume that a specific conditions are more probable than a single general one.

This is the sort of thing which college board exams have thrived upon, have built a billion dollar business on. And on these sorts of games, our "meritocracy" has been built. 

And you can say, well the guy who sees through to that answer, and gets it right while so many get it wrong is a smart guy and deserves to go to Harvard, but I beg to differ.

I would say the girl I knew in high school, who was so literal minded that when we were assigned, as a group, to explore the use of "roads" in "A Tale of Two Cities," she went through the book from page one to page 401, and found the word "road." And this was in the days before computers. She got into Harvard (then Radcliffe.)  The other three members of our study group in senior A.P. English class stared at her, stunned, as she rifled through her twenty pages of notes. Finally, I said, "Uh, Martha, I think what the real questions was how the idea of the road was used as a metaphor." 

She looked at me, vexed and said, "No, the questions says, 'Trace the use of roads through the text.' And that's what I've done."

"But, to what end?" I persisted. "Why you want to know that the word 'road' appeared on page 145?"

"Because," Martha said, exasperated with my inferior intellect, and my slacker ways, my unwillingness to do the slow, tedious thing, "That was the assignment!"

I have no doubt Martha got that feminist bank teller question (or questions like it) correct on the SAT.

What is a Symbionese?

I would have spent precious minutes during that test day dreaming about other attributes of the feminist teller: what sort of necklace she might be wearing under her bank teller suit, what her living room looked like.  Did she have something going with a lover who planned to rob the bank?  What would she have done if Patty Hearst had stepped into the bank with the Symbionese Liberation Army,  holding that assault rifle? Would she have lept over the counter and screamed, "I'm with you" ?  I'd have been day dreaming about that bank teller  for ten minutes and not getting on to the next question.

Not a bank teller, but way more interesting.

In fact, I would have argued to ignore the stuff they fed you about the bank teller and to say all we can know with any certainty or with more certainty about her is she's a bank teller is wrong. But no, the authors would say. The question is not really about her, it's about the probability that she is one thing being greater than the probability she is two. We are playing a game here and I have forgotten the rules, and made up my own game with my own rules to make the game more interesting.

Now, where was I? Is she a feminist?

The fact is, you will say, I don't belong at Harvard, and you are correct.

Stephen Jay Gould

But, then I tell you Stephen Jay Gould said, "I know the statement is least probable, yet a little homunculus in my head continues to jump up and down, shouting at me--'but she can't just be a bank teller' read the description.'"

Ah, now you are not quite so dismissive of me, are you?

A moment before Stephen Jay Gould weighed in, I'm just a loser who couldn't play the game. Now that I've got the big guy on my side, well, maybe I'm not so stupid.

And that's what I mean by the corruption of meritocracy.

When we have institutions filled with the "best and the brightest" we had better be sure we know what constitutes the best and the brightest. 

You may say, well Harvard was good enough at identifying talent to select Bill Gates and Mark Zuckerberg. But I would say actually, Harvard did not select them so much as they selected Harvard, because they could have gone anywhere and they thought Harvard was probably the best college. 

When we ponder how and why the sixty million plus losers out there came to reject the powers that be, the vaunted winners in the meritocracy of the American delusion, we might go back to the experience of those folks when they were sixteen, seventeen in high school getting told they could go no farther along the road to the top of the economic pyramid, because they did not make the cut, did not answer the questions correctly. 

Andrew Hacker

Andrew Hacker has argued the same thing in "The Math Myth" persuasively, persistently, patiently, but he's been largely ignored. Too many people are making too much money from a perverted system of testing for merit. As we used to say as premedical students, if calculus was not required to go to medical school they'd have to cut the math faculty by 60%. 

Somehow, the perversion of the selection process all along the line, from school days, to factory and industry and corporate advancement has finally caught up with this country, to the point where enough people have said, "I've got no faith in this system. Let's put a dummy like me in charge, and see if we actually do any worse."

I've got the best words. I come from good genes.