are their vocalizations. Because of vocal- if we really go in with the idea

that we evolved from something like the common ancestor of us and these creatures,

then maybe we can find some hints of our language in their vocal communication.

In the mid-1960s there was a very intersting discovery made; at the time, people thought

that, uhm, whenever a monkey gave a vocalization it was just sort of an involuntary

cry, like, "Ah!" you shout as you pull your hand back from the hot stove, or as

you're surprised as someone jumps out at you. And they don't usually use words-

sounds the way we use words, as voluntary signals that designate some feature

of the environment.

There are vervet monkeys in east Africa. They are small, grey monkeys that live in groups

of about 20 or so, and they were found, or thought, to give different alarm calls

to different predators. And the calls elicited different responses.

The monkeys would be out on the grass, and one of them would see a leopard and

give this distinctive leopard alarm call, and all the monkeys, all the other monkeys in the group,

upon hearing this call, ran up into a tree.

As if they had learned from the first guy's alarm call that there was a leopard nearby, and this caused them

to run up into the tree.

If an animal saw an eagle, and gave a very different alarm call, the other monkeys, upon hearing that

call, looked up in the air. Or they ran into a bush. But they didn't go up into a tree,

because eagles a very good at zipping through the branches and plucking off a monkey.

And if the first monkey to see the predator saw a snake, usually a python,

and gave this third kind of alarm call, the other monkeys in the group, upon hearing it,

stood on their hind legs and looked all in the grass around them.

Are the alarm calls really words? Are they really saying, "leopard"?

One possibility is that the alarm calls are just generalized alerting devices, sort of

like me saying, "Yo!" and then you don't know what I've seen, but you look up, you see the

leopard, and then you run up into the tree.

In that case they really wouldn't be like words very much. So we did an experiment

in which we waited until these vervets were foraging out on the- on the grass,

and then we hid a loudspeaker in a bush and played a leopard alarm call.

There's no leopard around, so it's just the call.

Everybody ran up into the trees.

If we played them an eagle alarm call, they looked up into the air.

And if we played them a snake alarm call, they stood on their hind legs and looked

in the grass around them.

And they really seemed to act as this- as if this call designated some feature of the

environment. And they responded according to their own vulnerability.

It- it's the first evidence of how you might, in a non-human animal,

get something that's very simple, it's not language, but it's a little bit like a word.

It's just the sort of thing that we might expect to find in an animal

with whom we share this ancestry. It's a sort of- of first gleanings of something

that might eventually become language.