Lastly, judging by his HN about page I wouldn't say his field of expertise is specifically mathematics, but rather that he has much broader educational expertise.
There's a book called "Why Don't Students Like School: A Cognitive Scientist Answers Questions About How the Mind Works and What It Means for the Classroom". The author explains (more or less) why Direct Instruction is so much more successful than other, newer, faddish techniques. Actually, Direct Instruction is also a bit faddish, and could certainly be improved by a more scientific approach.
Unfortunately, teacher trainers think that drill work is boring, and anything that recommends drill work must be wrong, because it is (to them) unpleasant. Maybe not as unpleasant as studying stuff you don't understand (which is what happens if you don't drill), but they don't have to accept reality.
And, under threat of punishment, once can't learn efficiently, anyhow.
For a fantastic critique of institutionalized schooling, check out Paul Goodman's Compulsory Mis-Education. http://www.amazon.com/Compulsory-Mis-Education-Community-Sch...
Recently I have heard that "Direct Instruction" is very effective, but I'm very skeptical as to what it is effective for. What exactly can you teach through that method, and what can you not? What does that method absolutely skip over? For example, through it, can you learn critical thinking, rhetoric/debate skills, rationality & logic, philosophy? Does it help students learn about themselves (self-discovery) -- how they think, learn, what their interests/passions are, what drives them? Does it help them find truth for themselves? Or, does it ignore all these and even more than our system today try to fit every personality into the standard box?
Does it at all care for each individual student or is it comprehensively uni-directional, from instructor to student?
IIRC, studies showed that DI did actually help the good students, the bad students, the mediocre students, the rich students and the poor students.
It also helped "higher order" skills.
I think that from a cognitive psychology perspective (though I'm not an expert), this is because the brain needs to learn fundamentals before it can work through more advanced things.
As for "self discovery", I think students are pretty good at that themselves. Current fads just pile on the homework though (because "testing doesn't teach anything"), which in my view is a big development killer.
I'm not a DI fan. I'm not a fan of capitalism either. But if you don't know why something works, you shouldn't just dismiss it for being unattractive.
I'm not sure what you mean by "student self-theories and various systems having to do with the existing educational paradigm", but axiomatic mathematics instruction leads directly into analysis, problem solving, and general logic; while that doesn't justify an educational curriculum consisting solely of mathematics, it does make it very important.