I am reading The Theory That Would Not Die by Sharon Betsch McGrayne. It is a book on Bayes’ Theorem. Bayes’ Theorem is simple but a bit too subtle to describe in a couple of snappy sentences in a blog post, so I won’t describe it here. You can check out the Wikipedia page if you want to know what it is.
The book is also not about the Theorem itself so much as the history of its use over the last 250 years since the Rev Thomas Bayes first developed a special case of it, and Pierre Simon Laplace developed the final general version of the theorem. The book also answers the question: What does it take to make politicians, generals and civil servants seriously combine data and maths to work out what the best course of action is, and to do it?
Answer: The prospect of mass starvation in the UK, and Nazi tanks driving through the streets of London. For example, before WWII, ships and planes searched for submarines as the captain and pilot saw fit. They would move in whatever search patterns seemed good to them.
Then came WWII. The UK was far from self-sufficient in food and early on in the war submarines were sinking ships bringing food to the UK far faster than new ships were being built.
Faced with this desperate situation, Bayes’ Theorem was used. Bayes’ Theorem can be used in a number of ways and one of these ways to take some data, and some guesses, and then produce a best guess at the optimal course of action. For example, if you know a submarine was say 20 km north and 15 km west1 hour ago, and can estimate the speed of a submarine, then Bayes’ theorem can be used to make estimates for where it is now most likely to be. And so logically you should start you search there.
This was done in WWII. As a scientist it is interesting to see this scientific approach to problem solving used in a different context. It is however a bit worrying that it was only resorted to in time of war. I worry that in peace time scientists who try and inject some science, some data, and some logical reasoning (with Bayes or otherwise) meet a less receptive audience. Prof Nutt’s attempts notably met with resistance.
