The thread had me leaning more to the pro-statistics side, if only because I was annoyed at the anti-statistics arguments. (For reasons similar to your comment.) But I didn’t stick in the Deming excerpt as any sort of representation of my belief, I was just amused at the synchronicity of coming across it while my computer was telling me to take a break from reading those e-mails. So I just posted it on the mailing list without comment.

Actually, my reaction to the excerpt was initially pretty close to yours: following on the quote you give, he says “His success would depend totally on his knowledge of the 14 points and of the diseases and obstacles, and the efforts that he himself puts forth”, which I translate as “drink my kool-aid or else”. Having said that, I actually do in general drink the Deming kool-aid (the book is quite good, or at least the third of it that I’ve read is), and it’s certainly important to understand the principles behind Deming (or lean or agile) rather than copying the surface manifestations.

So now I’m happy enough with the excerpt, but I still think there’s an interesting tension between the positive aspects of the excerpt and the negative aspects.

On my first reading I reacted quite negatively. It seemed to me that Deming was dismissing requests for _evidence_. And indeed “My answer to such enquiries is that no number of examples of success or of failure in the improvement of quality and productivity would indicate to the enquirer what success his company would have” sounds nearly like a wholesale denial of outcome-based evaluation of his method. (Replace “improvement of quality and productivity” with, say, “miraculous answer to prayer” and it is clearly an appeal to faith.)

Now, read specifically as an argument against attempting to learn solely from example, I would tend to agree with his point. (But I find it a bit amusing that the passage itself is fraught with examples!)