The sophisticated sounds of the simians.

While the laboratory education of Washoe the chimpanzee, Koko the gorilla and other domesticated creatures is well documented, scientists have assumed that apes and monkeys in the wild do not communicate naturally with any "language." There is communication, to be sure -- body movements in combination with various vocalizations are used to convey certain points -- but nothing, it was thought, approaching the sophistication of the sign and symbol language assimilated by the famous lab primates.

Now, however, anthropologists at the University of California at Los Angeles (UCLA) report that wild vervet monkeys have "vocal repertoires [that] are far larger than originally believed." Moreover, computer analysis of the monkeys' specific "conversational" sounds reveals them to be surprisingly similar in some ways to human speech, according to the researchers.

"It's like watching humans in conversation," UCLA's Robert Seyfarth told SCIENCE NEWS. The monkeys, he says, have "gone some way along the road to language." He reported the findings last week in Los Angeles at the annual meeting of the American Association for the Advancement of Science (AAAS).

Seyfarth and UCLA colleague Dorothy Cheney say that the "elements of language" they have discovered among vervets are much more subtle and sophisticated than the alarm calls given off by the monkeys when threatened by predators. The researchers had reported previously that vervets sound specific alarms, depending on whether they are threatened by an eagle, snake, leopard or other predator (SN: 11/24/79, p. 357).

"That's what led us to investigate their grunts," Seyfarth explains. In contrast to alarm calls, which are more like screams, monkey grunts are uttered in all types of nonthreatening situations. And, Seyfarth says, they all seem to sound the same. "Even experienced observers can't tell the difference," he says.

But after years of study in Kenya, Seyfarth and Cheney thought they may have heard tiny differences in grunts made by monkeys in four specific situations: approaching a dominant monkey; approaching a subordinate; acknowledging a leader's call to move onto an open plain from a sheltered area; seeing another group of monkeys approaching.

As they had done in their alarm call studies, the anthropologists hid loudspeakers in the natural environment of six monkey groups in Kenya and played grunts recorded in each of the four situations. They filmed the responses and found that when the listener was addressed by the recorded grunt of a subordinate it looked "sharply" and confidently in the direction of the loudspeaker; when addressed by a dominant monkey, it moved away; when hearing the "open plain" grunt, it looked out toward that area; when hearing the "other group" grunt, it looked out even more strongly.

After viewing the films, Seyfarth brought the corresponding vocal tape recordings back to the UCLA phonetics laboratory for acoustic analysis. He used computer software that conducts "Fourier analysis" of human speech. The process, which analyzes speech waveforms, revealed that the grunt waves from each of the four categories differed in two respects: the placement of the strongest energy and the change of energy peaks over the duration of the grunt. The latter, Seyfarth says, is similar to how human speech distinguishes between vowels.

"There are definitely some elements of language," Seyfarth says. "They are using sound to represent features of their environment." However, he notes, while monkeys appear to have semantics, they lack syntax. "They don't combine two or more [sounds], they don't make sentences and there is no particular order with abstract structure," he says.

Still, he says, the findings "illustrate that you can't judge the size of vocal repertoire by ear alone." And Seyfarth adds that his research opens up the possibility that other animals, particularly apes, may have natural communication systems in the wild that are far more developed than is now believed.