That erroneous point of view has led to many otherwise sensible people embracing the idea of a multiverse, but that’s the subject for another rant.

I think the failure to understand that quantum theory is (a generalisation of) probability theory has done worse. I remember feeling quite annoyed after reading Jaynes etc. and reflecting on what I’d been taught of probability and statistics as an undergraduate. I was [am] furious about the way quantum mechanics was [is] taught.

]]>This reminds me that I must invite John Skilling to Maynooth for a seminar…

]]>Steve and Anthony Lasenby also kicked off the Clifford Algebra revolution in Cambridge a decade later, but I think this one started with Anthony rather than Steve. The ‘Jaynes of Clifford Algebra’ is David Hestenes.

In those days there was much discussion of “objective Bayesian” and “subjective Bayesian”. And some people use ‘Bayesian’ to denote a way of finding a sufficient statistic – a concept which is as un-Bayesian as you can get. All of this verbiage, including the argument about the definition of probability, is got round by arguing that a number representing how strongly one binary proposition (“The data take value x”; “the hypothesis is true”)) implies another is what you actually want in all problems of this sort, and using consistency arguments to set up functional equations for this quantity – whose solution leads to the sum and product rules. Call the quantity whatever you like; it is what you need, and it obeys the sum and product rules.

Sir Harold Jeffreys was a Bayesian at a far harder time to be one, and it is a pity that he is remembered today more for his opposition to plate tectonics.

]]>