I spent the early months of 2016 in far, far western Oklahoma working on the early chapters of this book. Each chapter illustrated the main idea with a few examples from the literature, and one of the papers I had in mind for Ch. 4, "Vernacular Knowledge," was Goldin and Rouse's (2000) orchestra study. Their descriptions of blind auditions tied in nicely to their empirical analysis, I thought, and helped the reader understand viscerally what the data were showing impersonally.

But the empirical analysis. The more I looked at it, the more questions I had. Ultimately, these grew so numerous I pulled that text from Ch. 4--I couldn't hold up such a questionable paper. By summer, I was writing a much different discussion of the paper for Ch. 8, "Econometric Modeling": as an example of what not to do.

After my book was published, I learned others had traveled the same road at roughly the same time (2016-2019). Philosopher Sean Hermanson took the paper to task for the questionable assumption of exogeneity and numerous "odd patterns" in the data. Genomist (and far-righter) Jonatan Pallesen emphasized the imprecision and variation in the findings (as does statistician Andrew Gelman) and faults the media for reporting them too credulously.

That made four of us: a philosopher, a statistician, a genomist, and a low-status economist. And that is the point of this post. Economists of any stripe had more than fifteen years to speak up about the holes in the paper--and they did not. Just as in the movies, it took outsiders to bring critical faculties to bear. This is unacceptable in what is, putatively, a self-governing profession.