What did Descartes do for science?

Actually Lancaster University one Rag Week (infra-red shot)
According to Monty Python's Philosophers' Song, sung by the Bruces at the the University of Wallamaloo* (see below for the real thing):
Rene Descartes was a drunken fart: 'I drink, therefore I am'
However, Descartes tends to be held up as a scientist just as much as a philosopher. In Steven Weinberg's book To Explain the World which I've just reviewed, the author points out that while we owe a lot to Descartes' maths for providing the mapping between geometry and algebra, his thinking on the philosophy of science was more than a little shaky.

Specifically, Weinberg shows how Descartes, in his best bit of pure science, explaining how rainbows are seen at the angle they are, totally ignores his own method for 'Rightly Conducting One's Reason and of Seeking Truth in the Sciences.' According to this, Descartes says we should be highly doubtful about information that is derived from authority or our senses, but should instead rely on the power of reason alone. This was a kind of hybrid of Ancient Greek thinking and modern - he dropped the importance of authority and rejected teleology (where things are assumed to be the way they are because they are fulfilling a purpose), but he still wanted minimum observation and experiment, merely providing a spot of data to be worked with the power of sheer thought.

In practice, when working on the rainbow, he totally ignored his method and did something much closer to modern science. He guessed a mechanism and used that to work out the angles of incidence and refraction that would occur in a raindrop, finding they come close to what's observed. Then he does an experiment with an artificial 'raindrop' in the form of a globe filled with water and showed that the observed angles matched his predictions.

So, interestingly, at least once Descartes 'did' science in an effective manner, yet his philosophy of how science ought to be conducted was fatally flawed.

And philosophers wonder why scientists are sometimes a bit sniffy about their subject.

* I know this isn't the correct spelling, but it matches the picture