The birth of a [sign] language

September 18, 2004 4:02 PM

Experts Study New Sign Language System A new system of sign language developed by deaf children in Nicaragua may hold clues about the evolution of languages. When the country's first school for the deaf was established in 1977, children were not taught sign language but developed a system of signs to communicate. Childhood learning may determine linguistic rules ...They found that older students used hand signals resembling the gestures employed by hearing people, mimicking the entire event physically. But younger pupils - who had interacted with other deaf children from an early age - used a more complex series of signs. They split the scene into component parts and arranged these sequentially to convey the incident. The constructions resemble the way words and sentences are built in verbal languages, using segments structured in a linear fashion. This indicates that way the younger children learnt the sign language helped reshape it according to these linguistic rules.
............... Fascinating... /Mr. Spock
posted by y2karl (20 comments total)

The birth of a language Senghas asked signers of different ages to tell a story and found that, by the second 'generation' of children, those speaking the newborn Nicaraguan language had a similar system. Rather than have one sign for 'rollingdown', which would be the most economical option, they had two separate gestures. This suggests that children are born with a natural ability to break down language in this way. Hard wired rules like this help explain how language is acquired so easily.
posted by y2karl at 4:03 PM on September 18, 2004

Consider the languagehat signal lit. Would especially be interested in commentary vis-a-vis the role of small children in the formation of Pidgin and Creole Languages in comparison.
posted by y2karl at 4:08 PM on September 18, 2004

Glossary Of Terms Related To The Study Of Pidgin And Creole Languages from The Journal of Pidgin and Creole Languages
posted by y2karl at 4:12 PM on September 18, 2004

One of the most interesting things I studied in college was the chirology (aka phonology) of signs, particularly following Ursula Bellugi's work. The amazing thing is that just like spoken words are made up of distinct phonemes, signed words are made up of discrete, arbitrary elements - hand shape, motion, position, etc. What's surprising is how similar the linguistic structures are between spoken and signed language, given how different the modalities are.
posted by Nelson at 4:13 PM on September 18, 2004

Sigh.
posted by Ethereal Bligh at 4:15 PM on September 18, 2004

What's surprising is how similar the linguistic structures are between spoken and signed language

Depends which languages you mean. Western signed languages - ASL, BSL, NZSL - have a topic-comment structure, unlike the subject-verb-object structure of many languages such as English.
posted by raygirvan at 5:19 PM on September 18, 2004

Related thread.
posted by Gyan at 5:27 PM on September 18, 2004

Why the sigh Bligh?
posted by euphorb at 6:07 PM on September 18, 2004

What's surprising is how similar the linguistic structures are between spoken and signed language, given how different the modalities are.

For some reason, whistling languages come to mind.
posted by y2karl at 6:10 PM on September 18, 2004

I'm only an armchair linguist, but I've seen video of the Nicaraguan children's sign language (they've got snippets in the PBS Evolution video series), and it's one of the most beautiful things I've ever seen.
posted by Chanther at 6:27 PM on September 18, 2004

For some reason : NPR
posted by Satapher at 10:32 PM on September 18, 2004

What's surprising is how similar the linguistic structures are between spoken and signed language

Depends which languages you mean. Western signed languages - ASL, BSL, NZSL - have a topic-comment structure, unlike the subject-verb-object structure of many languages such as English.

Actually, what I think the original comment meant was that if you compare kinds of linguistic structure (linguistic universals) rather than particular linguistic structures in particular languages, it is (to many people) surprising to find that sign languages are the same as spoken languages. What universals there are do not seem to be specific to the modality of speech, or of signing for that matter. While ASL does not have a very fixed word order (what the poster called topic-comment), there are many spoken languages that have a word order determined by similar principles (what's more generally called information structure).

This is not surprising given that ASL is not related in any way to English. Given two languages that aren't related, we expect quite a few differences. But the general principles of syntax/phonology/semantics still hold.

By the way, languages that have a SVO order like english account for only about 42% of known languages (according to Comrie). This number probably includes some where the base word order is SVO, but is much less fixed than English. That is to say, most languages, and perhaps the overwhelming majority, are not like English in many respects.
posted by advil at 12:50 AM on September 19, 2004

euphorb: never mind. I'm all doped up. Don't mind me.
posted by Ethereal Bligh at 1:58 AM on September 19, 2004

I suspect EB's sigh may be related to the immense amount of effort and wordage he put into the discussion in the earlier thread Gyan linked to ("dear God, am I going to have to do that again?"), but I could be wrong.

At any rate, I will simply point to my disclaimer in that earlier thread: I have no expertise in either sign or child language, and I highly recommend Eldritch's informed commentary there, as well as advil's in this thread.

Excuse me, I hear prescriptivist nonsense is being spouted somewhere in Gotham City -- I have work to do!
*jumps into Hatmobile, drives off at reckless speed*
posted by languagehat at 6:17 AM on September 19, 2004

advil: point taken. But it's down to the semantics of what level of linguistic structures was meant.

The New Scientist comment was that putting symbols together in a "linear fashion" was a feature of all languages. Did they mean, then, that if you're saying something equivalent to "the cat sat on the mat", the symbols for "cat" and "mat" will always be strung together sequentially (in whatever order)? As opposed to a parallel fashion (say, signing "mat" with one hand and "cat" simultaneously with the other above it) that would theoretically be possible with a signed language?
posted by raygirvan at 6:28 AM on September 19, 2004

The Nicaragua case is hardly new of course, both Pinker and McWhorter mention it in their popularizing books.
I would like to point an error in the article. The year the deaf schools were founded was 1979, immediately after the overthrow of Somoza, by the Sandinista government. See also here, here and here.
posted by talos at 10:12 AM on September 19, 2004

And languagehat... I'm stirring up trouble here, but would you say that this case sort of strengthens the Chomsky's view of innateness of language? In the NYT article posted above (the last of the "heres") quotes Chomsky as saying:

Noam Chomsky, who calls what has happened in Nicaragua "a remarkable natural experiment," has for decades propounded the theory that there is a "biology of grammar" embedded in our brains. (It is no accident, he has argued, that every language from English to Zulu has subjects and verbs.) But he is wary of saying that Kegl's research settles the issue. "These children may have shown us something remarkable, if indeed they came up with this language with little or no input from outside," he says. "If that's the case, it's a very intriguing situation indeed."

Knowing your antipathy towards Chomsky's linguistic views, I wonder what your take is on this angle of the story.
posted by talos at 10:26 AM on September 19, 2004

My take is skeptical, but I'm glad to see Chomsky is showing an unwonted tentativeness in his reaction to this, which makes me feel less of my wonted eagerness to stick pins in him. I'm willing to go along with "a very intriguing situation." (And thanks for the links; I was just composing an LH post on this, and was glad to be able to add the NY Times Magazine story.)
posted by languagehat at 1:44 PM on September 19, 2004

The New Scientist comment was that putting symbols together in a "linear fashion" was a feature of all languages. Did they mean, then, that if you're saying something equivalent to "the cat sat on the mat", the symbols for "cat" and "mat" will always be strung together sequentially (in whatever order)? As opposed to a parallel fashion (say, signing "mat" with one hand and "cat" simultaneously with the other above it) that would theoretically be possible with a signed language?

I think the New Scientist was trying to distill (without understanding) a comment that had already been distilled for them perhaps several times. The usual claim is more like this (in a vague form): language differs from simple symbolic communication in that it has particular kinds of complex structures. These structures are usually characterized by general properties such as recursion (sentences embedded within sentences, noun phrases embedded in noun phrases, and so on). A linear structure wouldn't have this property. Particular theories of syntax have made much stronger claims about the kinds of structures that are universal, and as it's late I'm having a hard time trying to distill them down in a way that is more coherent than the New Scientist's comment, but perhaps you get the idea.

In fact, ASL does have cases of "parallel" pronunciation of things that in English would be multiple words. I don't know how widespread this is (I suspect it's not), but according to my sources certain manner adverbs are pronounced by mouth movements (approx. the english adverbs "carelessly" and "effortlessly") simultaneously with hand movements. What would be important is that these need to be represented by a complex linguistic structure, but this structure is "top-down" in some sense. That is, the adverb is contained in some complex phrase, which is contained in some other complex phrase, and so on. The particular ordering of adjacent constituents is already very language specific, so where the modality allows it, it's not surprising that adjacent constituents could be pronounced in parallel.

English has things that resemble this parallel pronunciation, at least with respect to other spoken languages. One of the things a pitch accent (and other prosodic contours) can mark is something called "focus" - think of a dialogue like "A: John had an argument with Mary. B: No, it was with JENNifer" with a "stress" on the first jennifer. This is one (common) kind of expression of focus, and it is effectively parallel. The pitch accent indicates focus properties while the word being pronounced at the same time has some other content. Some languages instead mark focus by having some morpheme that attaches to the focused phrase - not parallel pronunciation, of the same kinds of information that in English is parallel.
posted by advil at 1:43 AM on September 20, 2004

The Nicaragua case is hardly new of course
Indeed. It's only getting so much coverage now because of the study that has been published. (I wish I could get my friends to stop e-mailing me this story, I've known about it for years.)

a parallel fashion (say, signing "mat" with one hand and "cat" simultaneously with the other above it) that would theoretically be possible with a signed language?
It's theoretically possible, yes, in that signers have two articulators (each hand) rather than the one mouth used for speaking. It is simply impossible to produce two words intelligibly at the same time by voice. And yet it doesn't happen in ASL either; although it is be possible I have never seen it happen. Some of my high-school age Deaf clients have made a game out of trying to fingerspell CAT on one hand and DOG on the other, but that's about it. advil is right that mouth movements (also called non-manual markers, or NMM) add grammatical structure to produced signs, but they are not forming new words. The sign is always DRIVING but the NMM is what makes the difference between "a casual Sunday drive" and "road rage."
posted by etoile at 12:15 PM on September 21, 2004