Another interesting point; if Abelson were to write SICP again he would include a lot more on OOP than the short chapter it has in the book but at the time they didn't really appreciate the power of encapsulating state in that way.
It was surprising to hear one of the pioneers of functional programming say that. It goes to show that the founders of movements are oftentimes less rigid in their thoughts than their followers, indeed, the interviewer astutely brings up that SICP is highly regarded by people who perhaps aren't too fond of OOP. Hal Abelson really, really understands computation, so it shouldn't have shocked me to realise he's not dogmatic.