Once again I fell for the scam many reputable university presses run with their books in political science. Somebody has written a PhD thesis which contains a vast collection of correlation coefficients assembled from surveys. This data can hardly be of interest to anyone except researchers working on the exact same questions in the exact same field - that is, the same people who would read the thesis. But after a little rewriting, the university press publishes the thesis as a supposedly groundbreaking general work which "draws together insights from political philosophy, behavioural economics, psychology and anthropology".

That sounded like a book with lofty ambitions, so I bought it. It actually lived up to my expectations a little bit in the first chapter, which provides a versatile discussion of how and why status concerns can influence political behavior. This chapter contained at least a couple of smart insights that were new to me. But the insights come to an abrupt end when the author moves on from the "what I learned in my PhD work" chapter to the "let's look at the correlation coefficients" chapter. There's absolutely nothing worth reading in second half of the book. The author wades through all the boring experimental details of her PhD work without being able to develop her ideas at all. The only conclusion she deduces from all the data is that status concerns might sometimes influence politics - but this was already self-evident after the summary she gave in the first chapter!

I guess there's always some limited value in providing experimental support for a theoretical point, but proofs should be confined to academic theses. They should not be published as books aimed for a broad audience. Selling PhD-proofs as regular books might be profitable in the short term, but the prospective audience will certainly shrink in the long term. Academics will in any case download the actual thesis for free, and intelligent general readers will start to avoid books from university presses when they realize they are just thesis reprints with no general interest.