From Pushed by Jenifer Block (chapter 1, paperback pages 24-25):

Forceps are nearly museum relics, but the second stage of labor - the pushing part - is still actively managed by what's called directed pushing, or "purple pushing." For decades, women who weren't completely numb have been told to hold their breath and _Push!_ for ten full seconds, usually while a nurse counts out loud, a process that is repeated until the baby is out. In 1957, Constance L. Beynon, a British obstetrician, began doubting the ritual after observing a woman who birthed without incantation. It was accidental; she was attending a first-time mother for a colleague and tried to stall the delivery so he would arrive in time. "We ignored the patient's early straining efforts and when finally the head reached the pelvic floor, just allowed it to emerge slowly on minimal pushing, hoping every minute that her doctor would walk in," wrote Beynon. "The baby (8 pounds, 3 ounces) was born before the doctor arrived but with practically no effort on the part of the patient and an intanct vagina and perineum. The peacefulness and obvious ease of the birth were most impressive."

Further research revealed to Beynon that in women with physical disabilities or heart conditions, who were physically unable to push, "easy labor is remarkably common." Beynon then carried out a trial at Sussex Maternity Hospital with 100 first-time mothers and compared them to nearly 400 who were typically managed. For the study, the caregivers were not to suggest that the mother push; if they had to say the P-word, the case was recorded a failure. Of the 100 women, 83 delivered spontaneously within an hour and 98 within 2 hours, without any suggestion of what to do. Beynon wrote, "For many years now I have adopted the practice of allowing my patients to follow their own inclination in the second stage, forbidding all mention of pushing by those in attendance."

The female reporter who wrote this 300-page book has never been pregnant but she has read and talked to an impressive number of people. Her book is documented extensively. The statistics have an obvious slant, which is easy for a skeptic to notice (or someone who has read How To Lie With Statistics) but, overall, I think the author may be on to something.