Communication, language, and … apes

Got a letter from a reader.

In all of your spare time can you give me a detailed layman’s response to this question.

What is the difference between language and communication?

How is it that we know when an ape uses the sign for apple and the sign for want, even though he is communicating correctly, he is not processing the concepts in the same way a human would using language. Even if that language is the same signing language the ape is using?

I’ll make it as lay as I can.

You’ve hit on exactly the point that divides human language from animal communication — and it’s a distinction that people often miss.

There is a difference between language and communication. My cat can communicate anything he needs to — everything from ‘feed me’ to ‘let me in’ to ‘is there any more food’ to ‘don’t pet me there’ to ‘I don’t like this food’. But no way is the cat using language.

Human language is arbitrary. There’s no reason why ‘cat’ means ‘cat’, except that English speakers agree that it does. Pick a different language, and you’ll get a different word. And you can pick up the pieces ‘c’ ‘a’ and ‘t’, use them in other words, and there’s nothing catty about them. The sounds that come out of my cat, however, mean pretty much what you’d expect, and he never combines them to make new ‘words’.

Human language has syntax. My cat’s communication system is composed of parts whose meaning doesn’t change depending their relation to each other. ‘Meow hiss’ has the same meaning as ‘hiss meow’. For English, though, ‘John hit Bill’ is a different sentence from ‘Bill hit John’. And ‘Hit John Bill’ doesn’t have much meaning at all.

Human language is abstract. I can talk about things that aren’t in the room, or even things that are imaginary. Try that with an animal communication system. Even the celebrated European honeybee with its famous dance brings back a sample of pollen (stuck to it) to tell the other bees what kind of food has been found.

You can communicate with a look or a gesture, or even by not showing up somewhere. But language is a conventionalised system of symbols that we use to facilitate communication.

Now on to the signing apes. To understand what the ape ‘means’ by the sign for apple, we’d have to understand how meaning is represented in the brain, and then get into the ape’s brain somehow to examine it. Guess what? Can’t do it. (Come to think of it, that’s a bit of a stretch for people, too.) We can study brain activity, but how this relates to meaning is still highly speculative. For now, the ape’s mind is a black box.

So is the ape really using language, or just repeating signs? Prudence dictates that we assume less rather than more. And there’s good reason for this: people have been taken in before.

In Germany in 1899, there was a horse called Hans who could perform mathematical equations and language tasks by stomping his feet. Smart animals were nothing new at the time, but unmasking the fraud was never a problem. Just take away the trainer giving the secret signals and watch Bozo the Calculus Dog turn into a moron. Hans wasn’t like that. He could do his bit perfectly well without his trainer, who claimed to be as mystified as everyone else. One thing they did notice, though — Hans wasn’t good as answering questions if he couldn’t see the questioner. He also didn’t do well if the questioner didn’t actually know the answer.

What people figured out was that Hans was picking up on subtle cues that people give off unconsciously when speaking. If you asked Hans, “What’s two plus two?” you might lean forward a bit or raise your eyebrows. That was Hans’ cue to start stomping. When he got to four stomps, you might think, “That was it. Is he going to stop?” and you might straighten up, you might widen your eyes. Even a nostril flare was enough to signal Hans to stop. The horse was good.

Animal researchers have fallen for the ‘Clever Hans effect’. There’s also a good dose of wishful thinking and anthropomorphic fallacy in the brew. It’s a compelling illusion because we live in a world of people who understand us and respond. We therefore assume that non-humans are doing the same thing. Tell me the dog doesn’t look like he understands your English.

In the case of apes, researchers have noted that

• many signs are repetitions of what the trainer has just signed
• transcriptions of ape signing have sometimes been ‘cleaned up’, omitting repetitions, which makes the utterances seem more spontaneous and fluid than they really are
• trainers often give overly elaborate explanations of ‘what the ape meant’ by a certain sign

Linguists view ape signing as complicated tricks for treats. Apes have been taught to sign ‘want apple’ to get a reward. But they never sign ‘I like apples’ or ‘Have you got any apples?’ or indeed ever ask any questions at all. They may be using signs from a human language, but the evidence that this is similar to anything approaching human language use is not compelling.

The last word goes to Noam Chomsky, who said that discovering apes could use language would be like “suddenly discovering that humans could fly”. Humans don’t flap their arms and fly because they can’t. Apes don’t use language in the wild because they can’t. Communicate, definitely. Use language, no.

UPDATE: The Chomsky quote is a bit different than I remembered:

It’s about as likely that an ape will prove to have a language ability as that there is an island somewhere with a species of flightless birds waiting for human beings to teach them to fly.