Tuesday, June 24, 2008

What Is a Word, If a Baby Can Say It?

When my son Davy (who's now eight) was about ten months old he'd cruise around holding onto the couch saying "da da da da". He'd make the same "da da da" when he wanted to play with me. As an eager parent looking for milestones, I wondered if this was his first word but decided it didn't qualify. Soon he added "dis" to his reportoire: He'd point to something and say "dis", seemingly happy if I named what he was pointing at, but sometimes wanting more. What that a word? I asked an expert on infant language. Emphatically (almost tyrannically), she said no. Then more gently she added that developmental psychologists generally didn't take such seeming-words seriously until there were at least ten of them. Then they were words.

But ten of what? A hot trend recently in certain circles is teaching babies sign language: For example, if a baby makes a fist with one hand, that means "milk", if she brings her fingertips together, that means "more". Of course we don't want to be prejudiced against sign language: Words needn't be spoken aloud. The fist sign for milk is a word.

But now suppose that we have a psychologically identical case where instead of making a fist, the baby kicks her left foot in a distinctive way when she wants milk, and the parents learn to respond to that and reinforce it. Is that left-foot-kicking a word? Presumably we don't want to say that. To count such left-foot-kicking as linguistic seems to cheapen language too much. Ordinarily we think of language as something advanced, uniquely human or nearly so (except for maybe a few signing apes and Alex the parrot). The kicking doesn't seem to qualify, any more than we say a dog has language if he gets the leash when he wants a walk. In getting the leash, he communicates non-linguistically. So where to put on the brakes? Not every human communication is a word: A red stoplight is not a word, nor is a wink or a flag or a computer icon. (I think!) ;-)

Even very young babies will tighten their fists sometimes as a sign of hunger -- but surely that's not a word for a newborn. And babies seem quite naturally to point. Is pointing a word? My newly adopted sixteen-month-old daugher Kate will raise both hands over her head to communicate that she's "all done" eating. But just as the point might be a formalized reach, the hands up might be a formalized reaching up to be lifted out of her high chair. In fact, Kate has started raising her hands over her head in frustration when she has been trapped in her car seat too long and wants to be "all done" with that.

If a twelve-month old says "cup" for cup, we call it a word. If she says "mup" for cup, we find it cute and still call it a word. It seems to follow that if she makes regularly makes a sign language signal for cup, we should call that a word; and likewise it seems that if she makes her own unique sign for cup, we should call that a word too. But now the leash and left foot kicking seem to be back in.

So what is a word?


Anibal Monasterio Astobiza said...

I trust those motor theories of language that see manual gestures as the forerunner of language playing a key role the so called "mirror neurons", imitation and the like.

A word is any two-way simbolic mapping from the motor-articulary system to the sensory (abstract) world that allows the speaker or signier to use it or manipulate it (the motor-articulary system)irrespective of context.

Where to put the brakes? i think at the level of those motor actions or voice actions that cannot be iterated in a recursive manner to build nested structures because if not they are just simply one shot reactions to stimuli or emotions.

Anonymous said...

I think my hypothesis is in line anibal's comment (though not nearly as sophisticated). It seems we would have to be sure that a proto-language user is not merely working with a kind of stimulus response mentality. The infant might close its fist and trust that she will soon be drinking milk. But we cannot be sure that the infant realizes that the use of symbols is representing the milk. And I would think the successful use of symbols for representation is the hall mark of using a word (of course, successful use should not demand that the language user is reflexively self-aware that she is using symbols to represent). My hypothesis jives with the "ten word rule" because we would have evidence that the infant is not merely responding to stimuli in an "if conditions x obtain, then I y to get z" fashion and is instead in some rudimentary sense aware that her motions or utterances are picking things out in the world.

-Steve McFarlane

Anonymous said...

Hi Eric,

As for lot of other things, I don't think there is a clear sufficient and necessary features which make something a word.

I'm inclined to see linguistic communication as one of the practices in the community of humans. Once we see it this way, I think we should be inclined to think about the practice, not in terms of the simple 'is there'/'isn't there' distinction, but in the terms of slow development of the ability to a)become aware of the practice and b)take part in the practice. And, not just that, but as practice our communication has lots and lots of complexities. On the most "primitive" level we can use it for simple pointing, on somewhat more complex level for e.g. requesting something, further for informing, asking, and so on, to finally come to the level of telling stories, reciting poems, making puns, metaphors and so on...

So, I think asking if something is a word or not, is somewhat problematic question, and that we need to think just to what ammount the baby has become aware of the practice (able to figure out what some language-act is about), and to what amount it can take part in it.

Seems to me that asking if something is a word, is similar to asking if viruses are alive or not, in that that we are using a category which usually works on some pragmatic level, and try to apply it to a phenomenon in which the possible distinctions are much more complex than the used dichotomy.

Anonymous said...

"mirror neurons"? "iterated" motor actions? "two way symbolic mapping"??

And I though I was a slave to reductive neuroscience. Good God, you're not going to be able to talk about what a word is without mentioning, you know, a community of language users.

Do we need to go over the private language argument again?

Anonymous said...

Language is war by other means, and words are semiotic projectiles leveled at your adversary to make him dance ;-)

I'd like to go with what Anibal said, but I'm not sure how to take the "irrespective of context" part. Isn't recursiveness an abstraction based on the generalization of context?

Eric Schwitzgebel said...

Thanks for the comments, everyone!

Anibal: Will it follow from your view that babies don't actually utter words, since they can't build "nested structures" (i.e., grammar?) until well after their first (what we would usually call) one-word utterances?

Steve: If using symbols to represent is sufficient for word-use, then I wonder if computer icons or stoplights would turn out to be words, on your view.

Tanasije (and anon): Yeah, I'm definitely sympathetic to the broadly Wittgensteinian position you articulate. But another part of me (the part that wrote this post) still finds it a bit vague and unsatisfying. There seems to be some important landmark or boundary that is crossed toward language when the baby first starts to utter "mama" with the right communicative intent. I still feel the pull of seeing an important boundary here, though the reflections of my post, and thoughts of the sort you express, problematize it. Do you?

Cute remark, Badda!

Anibal Monasterio Astobiza said...

The acquisition of words by babies from the linguistic enviroment (parents, community etc.) is a process that involve a huge set mechanisms acting in concert during different temporal scales and frames in child development, but because i trust a motor theory origin of language is possible to assume that a given motor act (buccofacial gesture with sound with its underlying neural reprentation of conceptual content) could be intended to represent a given object in a straigthforward manner.
Something that we cannot say about other views.

In a chomskyan fashion they are in the adjusting and setting process phase of their "hardwired" principles until a full blown grammar cristalyzed.

The parallel is with learning to ride a bike or learning to walk with a complete cadence and biomechanics.

Babies first try to mantain
balance in a
trial and error and then hitherto produce a entire chain of upper extremities in sinchrnony with the lower extrmities.

Parsing a specific movement in the whole chain of movements for riding a bike, is a WORD.

If that movement (WORD) is not executed in the precise temporal order you cannot ride the bike (speak a language), so definitvely babies can utter single words as they can make a single motor act unchained to other motor acts.

Anonymous said...

anon here again.

In fact there are loads of different problems to choose from if you're going to give some complex neuro-based account of "words"... I just pulled the private language argument out of a hat.

For example, you're not going to get much of an account of reference just by looking into a speaker's head. ("Reason, Truth and History" is the most popular account of this argument, although it has a particular Putnam spin to it.)

...Unless of course, you want an account of word-hood that utterly divorces it from language. Judging from anibal's most recent post, that seems to be where he's going. It's not clear to me what value such an account would have.

Anonymous said...

Anonymous, I wonder if you are moving toward an account of wordhood that divorces it from brain activity.

Anonymous said...

now, now... even if you say that "linguistic communities" or larger social context matter to language (or knowlege, or ethics... ), that doesn't mean that neurological accounts don't contribute to the picture.

Saying the big picture matters doesn't on its own make you an anti-reductionist, any more than saying behaviour matters makes you a behaviourist.

Anonymous said...

That thing about 10 words and then it's a word may be based on trying to determine if a child has a language development difficulty.

There's no question your son had a word with "dis". My son did the exact same thing (his was "zhe ge"). In both cases, they were learned by imitating how the word was used by others. The use put to them is roughly the same as is used in the language in question. There's no doubt that's a word.

Da da da is just a vocalization. Babbling. A kid might mean something by it, but I think usually it's just parents reading into the da da da da.

My son as an infant had a special little cry he used when he was hungry. Maybe it was communicative, but I wouldn't call that a word.

We also taught him to make a fist if he wanted milk, I'd call that a sign not a word, just because that is sort of a convention, sign languages are languages that uses signs, and spoken languages are languages that use words. But that may be a distinction without a real difference.

Anonymous said...

Raising your hands for all done might be a Chinese thing, actually, I think clapping your hands, raising them in the air and saying "mei you le" is a common chinese child thing.

Also it's worth noting that A LOT of children learn to wave bye-bye before they have any spoken words. I wouldn't call that a word but I would call it a sign.

Eric Schwitzgebel said...

Thanks for all the interesting comments, folks! I'm totally wiped out from getting up early and staying up late a few days in a row now, for the meeting of the Society for Philosophy and Psychology. (It also doesn't help that I gave a talk and a separate poster and still have a commentary to give!) So, since infant language isn't my specialty anyway, maybe I'll use that as an excuse not to give much by way of reply. But shouldn't language have both a social and a neural component? And that last observation, Kenf, might explain why Kate picked up the signal so quickly and enthusiastically when I offered it to her (though I didn't do it in any knowledge of the Chinese game)!

Anonymous said...

Well, I have no specialty whatsoever, but my own feeling at the moment is that a word, if such a thing exists, may have more than a social and neural component. Each component represents a different vector of discourse that produces what a word is, so it would be closer to reality to say that we are talking about different things in each case, and that we only subsequently, if at all, combine them to form a supposedly fuller picture of a single object -- thereby, in effect, changing the subject. So I figure that there's nothing at all wrong with going all social or all neural with our account of wordhood, and the question "What is a word?" I take to be a strictly fun question along the lines of asking, "What shall a word be for us today?" The conversation could drag on indefinitely.

kvond said...

As to what is a word, I always found the idea forwarded by Rorty very helpful. One must have the capacity, in some sense, of saying "That looks like an X, but it really is a Y". This is "redescription" and at the highest levels it can be stated just like that, "All such Xs are really Ys". It is componented to language learning. But it also occurs in terms of behavior explanation. A young child, who cannot utter such a sentence, might understand that "Doggie" is not appropriate to this new object (a cat). Such "doggies" now become redescribed as "Kitty". Being able handle redescription seems to be qualifiable as word-use. Now animals signal specific conditions, and at times some seem able to handle redescription.

gregory said...

it is embarrassing to be in the company of people who think consciousness comes from meat, but this question of what is a word has a very elegant response in the world of sanskrit. the word for speech is vak and it relates directly to vibratory impulses of consciousness, which needs the understanding of gross and subtle to grasp, no western guy will touch this, panpsychism or not

so good luck following the grant money

Anonymous said...

But Gregory, weren't you born and raised among philistines in the United States? How did you become equipped with an understanding of such a foreign concept as vak? What does it mean to possess such an understanding (besides issuing vague pronouncements about its inscrutability to Western minds -- except, apparently, your own)?

Well, I do know that you've been residing in India for some time, so you've adopted its culture. But if vak is not translatable inter-culturally, how is it that it can, on the other hand, be translated intra-culturally between individuals? Maybe you're assured of your understanding only by the fact that no one offers to correct you in your use of the term, least of all with complete insolence.

If tomorrow some wisened chap comes up to you and says, "Gregory, I have to confess: you're understanding of vak has been flawed all along, but until now I've been too polite to correct you," would you debate him?

Anyway, good luck on your spiritual quest for...whatever.

Anonymous said...

So vak operates between individuals of the same culture, but not between individuals of different cultures. Unless you're from the East, then you can receive meanings from other cultures even if you can't transmit meanings from yours, whereas Westerners can only transmit without receiving. Unless you're Gregory Lent. But then how am I even writing this? Surely I must not be making sense.

Anonymous said...

Sorry, one last thing. Don't worry, I'm sure one will have greater luck "following the grant money" armed with proposals premised on meat consciousness rather than with an obscurantist New Age bricolage.

gregory said...

gosh, badda, i am so ignorant that i cannot understand what your replies are about, apart from emotional tone.

to the last one, yes, money is no correlate of wisdom

to the one prior, how does one know anything? what is knowing? who knows, or what is the knowing enitity?

the vak idea is based on the premise that there is no separation between what we call an I and what we call awareness, they are the same thing.

thoughts are considered condensations of this undifferentiated awareness, and words a further condensation, from subtle to more gross levels.

but i cannot expect to engage in any conversation that would be mutually meaningful for us, i simply lack the indoctrination in the concepts that professionals are burdened with, and so have no idea what, or even why, they talk like they do.

it just seems to me to be in the opposite direction from seeking wisdom or even of simply asking "why" and expecting an answer.

i can only offer than in some other cultures on this planet these subjects have been thought about long and hard, in terms of one's own experience, and the understandings gained seem to have a fairly valuable impact on applied awareness. enlightenment seems to be a real thing, as does "higher" consciousness.

Anonymous said...

Sign language with children is helpful way after infancy. When they are even after 2yrs old. When they have full use of language and are engaged in power struggles, sign language is an effective way to cue them to stop talking with words. It is helpful as well because they are motivated to learn more sign language so they can have a power struggle with you!. Any thing to stop power struggles and everyone save face. Annie

Anonymous said...

Not that there's anything wrong with emotion, right Gregory? But anyway, I wouldn't call you ignorant, though you do seem unable to recognize your own tone in other people's writing. No big deal.

Concerning vak, when you put it like that, I must say that I'm quite in sympathy with the concept. As a matter of fact, I do believe that we can cash out the idea of condensation in terms of indoctrination.

I'm not sufficiently indoctrinated into the meaning of wisdom to be too concerned about what I'm missing with respect to it.

But look. None of this means that we can't converse without exchanging meaning. It may not be the same meaning in each case, but so what? You posted here of your own accord, presumably to extract a bit of self-congratulation against the chatter of Western philistines, and I give it to you freely to do so because I'm more concerned with extracting new ideas which you inspire in me therewith.

Anonymous said...

Good Job! :)

Anonymous said...

Mostly because I just figured-out how to post photos and wanted to some of my OIF photoes online.