Bad Science And Paris

Bad Science

Some while ago I read a book called “Bad Science” which put the boot into alternative medicine and the cosmetics industry; and loved it. There's an associated website. Now, by chance, I've found another website I love which is dedicated to bad science or, more specifically, to the dangers inherent in coming to any conclusions based purely on mathematical correlations. It's TylerVigen.com.

Did you know, for instance, that there is a very close correlation (0.94) over a decade between cheese consumption per capita in the USA and the number of people who die each year by becoming entangled in their bed sheets? If you're going to the USA (or already there) better not eat cheese before going to bed unless you remove the covers. Even worse, margarine lovers who marry in the US state of Maine are virtually certainly doomed to misery; there's a 0.99 correlation between margarine consumption and the divorce rate in the state. As you might guess, the site simply looks for graphs which match over a significant period of time on any subject and calculates the correlation coefficient; you can draw conclusions if you want to but probably only a politician would.

I loved this site particularly because it reminded me of a tongue-in-cheek article by Michael Frayn in his Miscellany column in The Guardian, decades ago. At the time, the populist newspapers were making a scandal out of a fashion among some schoolgirls to advertise the fact that they had lost their virginity by wearing a golliwog broach. The broach was obtained by sending some labels off jars of a popular brand of marmalade, which had a golliwog as its brand symbol, to the manufacturer. Frayn found a graph of sales of the brand of marmalade and also a graph of schoolgirl pregnancies and saw that they matched. The conclusion the article came to (tongue in cheek) was obvious. If schoolgirl pregnancies were to be reduced, contraception and sex education were irrelevant; what was needed was to reduce consumption of that brand of marmalade.

Paris

I'm just back from two days seeing old American friends Ed and Jeanne in Paris. I hadn't seen them for a decade and, in the meantime, Ed had contracted Alzheimer's. There were obvious problems but we were able to reminisce happily over drinks and meals. Ed did occasional seminars in London over a period of years and had obtained consultancy assignments for me at the State of Oregon and Nike.

An item on Jeanne's agenda was getting an ice cream at Berthollin's on the Ile St Louis, which we did, and both also wanted to visit the Musee Rodin again. That was a disappointment as there were few of Rodin's works on display. I understand that some of them are probably on loan from time to time but too much was missing. The Kiss and The Thinker were there, inevitably, but the bronzes of The Burgers of Calais that I had seen previously there in the grounds were missing as also were all the dancers and two of my particular favourites, The Cathedral and She Who Was Once The Beautiful Wife Of The Helmet-maker. The Cathedral, for me, is both a beautiful piece of sculpture and a beautiful concept: two hand in concave shapes with the fingers just touching. The whole museum had had a make-over since I was there before and the pieces of sculpture that were there were displayed in acres of space, which I suppose was to their benefit. However, English translations of the French titles had been added and obviously not checked by any native English speaker. Millions must have been spent on the refurbishment; wasn't there enough left to check the translations? Why do the French persist in doing this? I'm beginning to think it must be some kind of revenge for Waterloo.

If the Musee Rodin was a disappointment our hotel wasn't. The Hotel Chopin is at the end of a quiet passage right in the heart of Paris, off the Boulevard Montmartre. It's a small old-fashioned hotel with tastefully decorated rooms, excellent service and very modest prices. This last applies to the train journey also; the 500kms from Avignon to Paris (in 2 hours 40 minutes) cost me about the same as the 30 miles from Reading to London would have in England.