I find myself of two minds on this.
The theory seems almost bizarre on its face -- if smarter parents with big-brained babies were less likely to survive reproduction, natural selection would automatically select for smaller heads. Because, you know, the mothers of big-brained babies would die during childbirth more frequently, which (by definition) means reduced fitness for reproduction. It then comes down to whether small headedness or adult intelligence is the larger predictor of successful adult reproduction, I guess.
It's always a bit of a "just so" story to try and reason out the precise mechanisms of long term statistical natural selection that is influenced by MANY factors, but this one seems a bit more ridiculous than most.
But also, "the obstetrical dilemma leads to a widespread notion of the female body as inescapably defective"... seems like a silly takeaway. There may be ample scientific problems with the obstetrical dilemma, but worrying about the message it sends is a moral position, not a scientific one. Acknowledging that certain activities -- such as pregnancy and birth -- carry statistical medical risk is not an accusation that anyone is "inescapably defective". That's an extremist, normative interpretation of the medical facts.
If a doctor tells somebody, "you have a medical condition that is statistically likely to increase risk of X", that's NOT telling somebody that they are "defective". There is no active creator who "made" a person, there was no decision made to produce an inferior product, there is no fumble-fingered worker who screwed up. The recombinative genetic lottery is what it is, and a plain statement of the facts is not a value judgement.