Nudge: Improving Decisions About Health, Wealth, and Happiness,
by Richard H Thaler and Cass R. Sunstein,
Yale University Press, 2008.
One of the themes of the novel Bend Sinister (1947) is the principle that Tyranny & Competence are mutually exclusive concepts. That thought has given me a lot of comfort over the years. For example, very little harm (and some harmless fun) has come out of America's efforts to run propaganda out of the State Department over the last few years.
Recently, however, with the advent of Professor Cass Sunstein as the Administrator of the White House Office of Information and Regulatory Affairs, things have started to look a bit grim. Indeed the prospect of combining choice architecture with the quality control of government department statistical products is downright alarming.
Some time ago Sunstein's name started to come up in the blogosphere so I borrowed his book Infotopia (2006). That text was only vaguely disturbing, but then I read the online 2008 academic paper he co-authored on fighting conspiracy theories, which was very disturbing.
So the other day my turn came up to borrow Nudge from the library, and I feel a lot better now
Things didn't start off too well. Thaler and Sunstein get the ball rolling in the Introduction with "Carolyn, … the director of food services for a large city school system." (1) It's not until two pages later that they casually point out (if I'm paying attention) that Carolyn is in fact bogus. This (along with a subtle clue they slide into the first sentence) is a level of deviousness that is not, IMHO, really fitting in non-fiction. Bogus is OK. I was always fairly comfortable with — Fred (not his real name) an architect (not his real job) … — from the Readers Digest in the '50s, but don't really like playing mind games with my authors when I just want information.
But their opening gambit in Chapter 1 is sheer joy.
Have a look, if you will, at these two tables: (17)
Suppose you are thinking about which one would work better as a coffee table in your living room. What would you say are the dimensions of the two tables? Take a guess at the ratio of the length to the width of each. Just eyeball it. (17)
Now as any child of a species that has been throwing rocks and hitting things for the last several hundred thousand years will immediately see, the one on the left has ratio about 3:1, the one on the right about 1.5:1, just like the "[t]ypical guesses" (17) cited in the text.
It's easy to see that the viewer of the above picture is located above the two tables and looking down at an angle. As it happens, Doom North has a coffee table very similar to the left-hand one and after some careful inspection Friday night I determined that I was looking down 45 degrees from the horizontal to achieve that effect. That means the vertical axis in the picture is foreshortened by about 29 percent. Since the picture-ratio of both tabletops is 2:1, the tables as pictured then have ratios roughly 2.5:1 (left) and 1.6:1 (right), not too far away from those typical guesses.
By the way, the foreshortening effect is simple to demonstrate. Just get a piece of graph paper and look at it from an angle.
What's really delicious is the date on that picture. It would seem that at least a smattering of cognitive scientists believe the two tables have identical shape and have been clinging to their error for up to a generation, in spite of a steady stream of subjects giving them correct responses. And Thaler and Sunstein, with all their editorial help, fell for it too. In spite of the fact that their entire Chapter 3 (53ff) is on peer pressure.
So I don't think we have to worry very much, at least until Cass gets the memo about the cosine of pi over four being 0.7071. And I suspect it's going to be a good long time before anybody at OIRA gets up the courage to do that









