WASHINGTON - Every parent is familiar with "motherese," the slow, high-pitched, singsong tone that mothers all over the world use to talk to their babies.
You might suppose that this manner of speaking is just a simplified, degenerate form of grown-up language.
To the contrary, some scientists think motherese might lie at the root of human language - long ago when our primitive ancestors came down from trees and started to walk on two legs.
Dean Falk, the head of the anthropology department at Florida State University in Tallahassee, calls it the "putting the baby down hypothesis."
As they began to walk erect, instead of on their knuckles like chimps, early members of the human family developed taller, thinner, less hairy bodies than their apelike predecessors. The female pelvis narrowed as infants' brains were getting bigger. Babies had to be born early to squeeze through the birth canal.
These helpless creatures couldn't cling to their mothers' fur or ride piggyback the way 2-month-old chimp youngsters do. They had to be held in their mothers' arms.
But mothers foraging for food needed to park their babies on the ground so they could use both hands. While they were busy, they emitted soft, cooing sounds to keep the infants calm and quiet, so as not to attract predators.
"To feed yourself, you have to put the baby down right there next to you," Falk said in an interview. "You have to keep the baby with you or it'll be eaten by a lion. You don't have touching contact, but you can reassure the baby with melodic tones. It's a substitute for cradling arms."
Early human mothers who behaved this way raised more children successfully than those who didn't, Falk theorized. Their genes were passed on to their descendants and became a universal human trait.
"The urge to go 'ga-ga, goo-goo' to small babies is universal," she said. "This became a genetic pattern. If mothers didn't vigilantly attend to their infants, their infants were lost, and the mothers' genes were lost to the future."
Falk acknowledged that her theory - to be published in the December edition of the journal Behavioral and Brain Sciences- is controversial.
Some researchers say the idea is plausible but unproved. Some say it might explain the importance of vocal sounds to foster social and emotional bonds, but not something as profound and complex as language. Others reject her position.
"We believe that caching [parking] a baby would be a danger too great in human prehistory, and thus could not serve as the basis of pre-linguistic vocalization," Nicholas Thompson and Rosemarie Sokol, psychologists at Clark University in Worcester, Mass., wrote in a commentary on Falk's theory. "Rather, infants were most likely carried at all times."
Sherman Wilcox, the chairman of the linguistics department at the University of New Mexico in Albuquerque, was supportive.
"Falk presents a compelling argument for the role of motherese in the evolution of language," he said, "weaving a range of scientific evidence from anthropology, linguistics, neurology and psychology into a solid case for her 'putting the baby down hypothesis.'"
"It strikes me as fairly plausible," said Steven Pinker, an eminent psychologist and language expert at Harvard University in Cambridge, Mass.