« Depressed Reflections | Main | Cholera In Iraq »

September 12, 2007


Animals, some animals, can think and feel; the evidence is almost overwhelming. One thing they don't do, not even close, is language. Not even Alex: words aren't language. (You didn't say otherwise, but others have, and have pointed to Alex as proof.) But their utter lack of that specific skill doesn't mean they don't have other, arguably more morally relevant ones.

Damn, I wish I were that smart.

One thing they don't do, not even close, is language.

Actually, it's been demonstrated with boring predictability that if you expose a young chimp to language use of a kind the chimp can learn (gestural, visible) the chimp does in fact acquire language - never above the level of a human toddler, but definitely language. If you think about it, it would actually be surprising if our closest primate relatives didn't have a similar ability to our own to acquire language by being exposed to language.

Language use in other animals is more debatable because, whatever it is, it's not going to be quite like our language use. But with chimps, it's definite.

He produced instead each of the wrong possible answers from the appropriate category, repeated each wrong answer, then grabbed the tray liner and tossed all the objects to the floor.

aha! Bush is a parrot!

Ah, the Norwegian Blue. Remarkable bird.

DPU, Language Log used "This parrot is no more!" as the headline.

I didn't want to be that obvious. Out of respect.

"I'm not sure the astonishing nature of Alex's language-learning comes through here."

agreed. you really have to see the video.

i remember very clearly the sequence where they say good night to each other, with some variation of the quote from that article, sc. “You be good, see you tomorrow. I love you.”

because the parrot has mastered intonation, it sounds like it means what a human would mean when saying all that.

but that's preposterous. as christopher m says, alex just didn't have language; he did not have anything like the generative ability to construct the infinite number of sentences with those primitives that we could (e.g., 'since i love you, it's good that i will see you tomorrow', "i love it that you are good", etc.)

but if you have had kids, and gone through bed-time rituals with them, then you know that bed-time endearments are pretty rote themselves. little kids learn to say 'i love you' in response to their parents far earlier than they can be said to know what they hell they are saying or what it means.

they just know that it's what you do now, with this person, in this sequence, with these feelings. for little kids, it's just parroting, but it's parroting, with deep feeling.

it's not obvious that a lot of grown-up language, esp. the expressions of deep feelings, are much different from that.

so: it would be a mistake to watch alex and irene say good night to each other and conclude that alex is a language user.

but it is far harder to shake the sense that they are doing all or most of what any parent and child do when they say good night to each other.

the more morally relevant stuff.

good night, alex.

Alex could verbally identify objects by shape, color, and number.

I have known African Grays who could ask for their favorite foods by name, who knew what they were saying. I've personally seen and heard this.

How is that not "using language"?

How is that not "using language"?

Using language is what humans do. Therefore, when other animals use what may look like language, it obviously isn't language, because animals don't use language. You know, like animals don't use tools because only humans use tools therefore when an animal does something that looks like it's tool-using behavior, obviously it's not.

A lot of this "logic" comes from the assumption that there is a sharp dividing line between humans and all other animals, which clearly makes humans absolutely and completely different from all other animals. I'm not sure why this should be thought to be necessary. *off to pick more nits*

the question you need to ask, when thinking about extraordinary animal behavior, is does it entertain us.

I believe the idea that a parrot's grasp of words is not language comes from the more in-depth definition of language that, say, a linguist would use. Not that linguists agree on any damn thing, but generally the ones who say that only humans have language require language to consist of verbs, nouns and adjectives, in a structure that can be used to say anything you can think of. like kid bitzer said, you have to be able to switch words around in order based on the meaning you're trying to convey. my old linguistics prof would say that parrots have no sense of pragmatics, nor probably of syntax.

on the other hand, if you view language simply as the ability to convey an idea with words, then clearly some parrots can do this, although they may be extremely limited in the ideas they can communicate compared to humans.

A lot of this "logic" comes from the assumption that there is a sharp dividing line between humans and all other animals, which clearly makes humans absolutely and completely different from all other animals.

But haven't you heard, we're unique in the universe, made in the image of (and thus favored by) God. A fair amount of arrogance there.

I like the idea that God is a grey parrot.

if you view language simply as the ability to convey an idea with words...

With all due courtesy and so on, WTF else is language supposed to be, other than 'the ability to convey an idea with words'?

I think what you're actually trying to say is there is, or may be, a difference in the cognitive level achievable, and expressible, by humans versus non-humans. That Alex, prodigy though he was, was still never going to be able to launch into a Socratic disquisition, much less a Maslovian one; that the abstract notions of self-actualization, authenticity, etc., are beyond a parrot's cognitive reach, and therefore beyond its linguistic reach.

And that I might grant you - might, that is, if we were sure that the capacity to know/intuit such cognitive leaps was equal to, or even dependent on, the capacity to express those cognitive leaps... and, moreover, to express them in some other species' language.

So, actually, I'm not at all sure I can grant you that. Alex - or any comparably intelligent non-human creature - might certainly be able to know/intuit higher-order concepts that they can't express to us. In fact, I really really hope the language studies with parrots continues - precisely because, at some point, it might in fact be possible to teach parrots enough intricacies of human English speech for them to tell us what goes on in their heads.

"Even if a Lion could speak we wouldn't understand what he was saying" Ludwig W.

With all due courtesy and so on, WTF else is language supposed to be, other than 'the ability to convey an idea with words'?

Tayi is right that linguists agree on very little, but one thought that it's actually the ability to spontaneously convey new ideas. Another is the ability to create new symbols to represent new stimuli.

The problem with proving Alex is 'using language' in the same way a human is is the same that one faces with the Turing test, in that the mere appearance of well formed strings in response to stimuli doesn't necessarily indicate the presence of an active intelligence (or does it? ...). It is true that some linguists like to draw a bright line between humans and other species, but at some point, a difference in degree becomes a difference in kind. Pepperburg herself suggests the same in an interview here

What the data suggest to me is that if one starts with a brain of a certain complexity and gives it enough social and ecological support, that brain will develop at least the building blocks of a complex communication system. Of course, chimpanzees don't proceed to develop full-blown language the way you and I have.

This is probably a statement that most linguists would accept.

CaseyL, I do agree with you that pretty clearly, once you start talking about organisms that can communicate an idea with words, you're talking about language, and from there the discussion should be about what kind of language it is. Parrots almost certainly view the world in a way completely alien to human thought, but I don't know what else to call what they do. Its language to me, although like I said, my linguists prof would disagree.

I think it's possible to just sidestep a lot of the problems about whether Alex could use language by just talking about the specific capacities he did and did not have. He did generate new symbols, as did some of the signing chimps; and he had some grasp of syntax, fwiw.

Haven't read the whole thread, maybe this has been posted already, but: There is a great video of Alex that you can see at Scientific American Frontiers -- it is at the bottom of that page, misleadingly titled "Animal Einsteins". Also some nice footage of teaching chimps arithmetic.

It's also interesting that chimp language experiments seem to indicate that there is a relationship between the theory of mind and language use. Chimps taught sign language (to which they are more suited than speech, owing to the shapes of their mouths) have a more sophisticated theory of mind than those not taught.

I once knew a particularly stupid albeit good-hearted large dog (I privately called him Lenny). My friends and I used to imagine his cognitive processes going something like "hissbuzzhissbuzz stick? hissbuzzhisshiss stick! hissbuzzhisshiss stick . . . fetch? buzzhisshiss fetch!" and so on.

It strikes that, from Alex's point of view, his understanding of the situation might be that he taught a human to give him grapes.

Thanks -

This morning when I was watching that video, I was trying to think about what Alex was doing as correlating symbols with objects (and stuff like "color" and "material" which aren't quite "objects") and wondering how much of an understanding he had of the process of assigning symbols. I found it utterly enchanting to watch him and Dr. Pepperburg, whatever it was that they were doing.

Dogs speak dog. People speak people. My dog speaks to me all the time. Just because I don't know what she's saying most of the time doesn't mean she not saying it. As a matter of fact, she understands me way more than I understand her. So, who's smarter?

but if your getting fed, receiving toys, and your opportunity to use the bathroom without getting smacked depended on understanding your dog, you would probably understand it better.

A nice little article on dogs' understanding of human expression here.

JakeB, sort of like this?


actually, I did think of the "What dogs say" cartoon, where you see a whole neighborhood of dogs all going "Hey! Hey! Hey!", but didn't find it quickly enough to suit my laziness.

What's interesting to me here is the association of thought with language.

People think in lots of ways that are not expressed in language. Music, dance, sculpture. The intuitive thought process that you use to decide what seat to sit in on a crowded bus.

All of these things can be talked *about*, or explained in words after the fact, but as thoughts they aren't, and often can't be, uttered as language. But, thoughts they are.

Then, there's simple intuitive awareness without thought. Not common in our busy lives, but it's there.

I'm not sure how animal awareness presents itself to the animals in question, but if you watch animals, even the "less intelligent" ones, it's apparent that they reason, make choices, and have complex and fluid interactions with each other. That seems like intelligence to me.

One thing that's always seemed to me to be a marker of higher intelligence is play. By play, I mean activity that doesn't have a survival or adaptive function, but is something an animal does just to entertain itself. Crows do it, dolphins do it, some primates do it. Dogs will play until you wish you would never see a tennis ball again as long as you live.

I used to have a parakeet that would alleviate the insufferable boredom of life in a cage by dropping a feather and watching it flutter down, over and over again. I also had zebra finches, and they never did anything like that. The parakeet was bored, and made up a game to amuse itself. Intelligence is required for that.

Thanks -

Modesto Kid: besides Alex, your link had footage of one of my favorite animal intelligence studies ever. (I had forgotten where I saw it.) In it, Sheba, a chimp who knows about numbers, is shown two bowls. Whichever she points to first, the contents (or: the relevant number of candies) go to another chimp and the content of the remaining bowl goes to her, so it is in her interest to pick the bowl with the smaller number.

The cool thing is: she can reliably do this when the bowls contain numbers (e.g., a plastic two); but when they contain actual candies, she always picks the bowl with the bigger number. The theory is that when a mere abstract representation of the number is present, she can pick the bowl with the smaller number, thereby getting the most candies for herself; but when the actual candies are physically present, she can't help herself and has to reach for the bowl that has the greatest number of them, even though (as her performance with the numbers shows) she understands how the game works.

I think this is fascinating.

Yeah, I liked that a lot too. (And Sylvia, when I got home this evening, wanted immediately to watch the parrots and chimpanzees video a second time.)

As far as I can see the parrot's abilities were:
1. recognizing abstract categories (colour, shape, number)
2. negating those categories (i.e. noting their absence)
3. combining those categories
4. expressing that understanding in an alien code system (human words)
From my (layman) perspective 2&4 are the most remarkable achievements both transcending "parrot fashion".
Many IQ test for humans contain tasks of the same kind.
I think the important step is from simple event triggering by code word ("Polly wanna cracker" = "feed me") to the deliberate selective order (i.e. specifying what type of food is wished at the moment). the latter implies a synthetic approach that qualifies as "language" in my book.
The fine distinction would be, whether the animal in question is able to produce a signal that it had not learned in that precise version but only in separate parts. If it is able to combine, it is speaking, not just reproducing. The advantage of abstract vocal answers as opposed to a "multiple choice" system is that the person conducting the experiement is far less able to influence the "answer" (as in the notorious Clever Hans case).

I already knew parrots and other birds have more cognitive powers than Conservatives.

No big revelation there.

"He knew his colors and shapes, he learned more than 100 English words, and with his own brand of one-liners he established himself in TV shows."

At first glance I thought this was a piece about Bush's Iraq speech.

Alex is in the NYT again (Week in Review):

In an talk on Edge.org, Dr. Pepperberg told of an effort to teach the parrot about phonemes using colored tokens marked with letter combinations like sh and ch.

"What sound is green?"

"Ssshh," Alex answered correctly, and then demanded a nut. Instead he got another question.

"What sound is orange?"


"Good bird!"

"Want a nut!" Alex demanded. The interview was over. "Want a nut!" he repeated. "Nnn ... uh ... tuh."

Dr. Pepperberg was flabbergasted. "Not only could you imagine him thinking, 'Hey, stupid, do I have to spell it for you?' " she said. "This was in a sense his way of saying to us, 'I know where you're headed! Let's get on with it.' "

Fine distinctions aside, it seems to me Alex had a sense of humor.

The comments to this entry are closed.

Blog powered by Typepad