Hmm, based on the people I work with, about 90% of the population! Actually, the people in our lab are almost entirely not PhDs, which makes us very unusual, and the only people in the entire group (other than the secretarial staff, which is 2/3 of a person) who aren't PhDs or postgrads. We have a single PhD out of eight people, and shes doing work which is more related to science than the rest of us, who are really doing engineering.
I'm not sure I agree with that rant in the context of the group I work in at Imperial. PhD failures are pretty rare, because they are extremely embarrassing for the supervisor. If someone fails their viva, there are going to be questions asked of their supervisor, since the student should never have been put forward for a viva until their thesis has been written to a suitable level, and they understand everything in it, and the relevant background.
I also found the statement "...Bassnett once spent 10 hours correcting a single chapter of a student's PhD, a large part of it for grammatical reasons..." odd. If anyone I know was sitting on a viva panel, and found the thesis was like that, then assuming that the scientific contents was OK, the student would be sent off to rewrite the thesis. Virtually no one ever passes a viva without minor corrections, many of which are grammatical and spelling. No one is going to correct the thesis for them, although they'll possibly mark the errors.
I'm guessing that PhDs in areas wildly different from Physics aren't dealt with in the same fashion, which I guess one of the things that the rant is complaining of, a lack of consistency.