I know: “Dog Bites Man”. But I think it’s symptomatic.
In The Nature of Physical Law, Feynman writes that Olaus Roemer, “having confidence in the Law of Gravitation”, concluded that discrepancies in the times of observations of eclipses of Jupiter’s moon Io were owing to the fact that the velocity of light is finite (and so takes longer to reach the Earth when the source is farther away).
Roemer’s results were reported anonymously to the French Académie des Sciences in 1676. A manuscript note cited in the paper below shows that Cassini too had arrived at the view no later than August 1676.
Newton’s Principia was published in 1687, over ten years later. Roemer could not have had confidence in a Law not yet published.
It’s an elementary mistake. It could have been corrected in a few minutes. But scientists often permit themselves to commit errors on historical matters so grievous that, were they to have committed analogous errors on scientific matters, they would be utterly embarrassed.
Feynman likewise gives a misleading account of the discovery of Neptune. He writes that after John Couch Adams sent his proposal to the Royal Observatory,
‘How absurd,’ said [the Royal Observatory], ‘some guy sitting with pieces of paper and pencils can tell us where to look to find some new planet’ (Feynman 14).
Feynman’s version makes a good joke, but it wrongly implies that Airy, the Astronomer Royal, rejected out of hand Adams’ calculation of the orbit of the new planet, and worse that Airy would not have searched for new planets on the basis of calculations. In fact, as soon as Airy read the just-published paper of Le Verrier, Adams’s French rival, in late 1845, he rushed to make the relevant observations.
Feynman’s book had at least one edition after it was first published in 1965. The edition I’m reading is the 1994 Modern Library edition with an introduction by James Gleick. Gleick’s companionable introduction makes no mention of mistakes by his “rough-hewn, American” author. I don’t suppose that doing so could be demanded of him.
One could argue that the book is a classic, whose mistakes should no more be corrected than Joyce’s when he implied that the Sundam Trench is over 8000 fathoms deep (it’s less than half that deep). But Ulysses is fiction. Feynman’s book is not, and readers are likely to accept on his authority its claims in history, not just in physics. Gleick might not have felt it necessary to correct those claims himself, but he or the editor at the Modern Library could have had someone do it. There’s no point, though admittedly there is profit, in perpetuating falsehoods.