Language of Numbers in Nicaraguan Sign Language

February 10, 2011 12:41 PM

Nicaraguan Sign Language is a unique language, created by school children in the late 1970s and early 1980s, who previously had minimal success at being taught to lip-read and speak Spanish. This community has been studied as an example of the birth of a language from its beginning (PDF). A recent study has investigated the ability for those who speak Nicaraguan Sign Language to express exact, large numbers. Unlike the Pirahã people of the Amazon (previously) who may not have the need for specificity in large numbers, the deaf in Nicaragua are surrounded by a culture that interacts in specific numbers, yet it appears they lack accuracy with numbers higher than three or four.

Elizabet Spaepen of the University of Chicago was the lead author of the recent study, entitled Number without a language model, which was also covered on ScienceDaily.

posted by filthy light thief (21 comments total)

Oh man, I remember hearing a story about this on... either TAL or RadioLab. Aha! It was Radiolab.
posted by kmz at 1:02 PM on February 10, 2011

One thing I wanted to point out is that the participants in the study (I just read the study) are described thusly:

Their hearing losses prevent them from acquiring Spanish
and they do not have access to Nicaraguan Sign Language.

Homesign (in Nicaragua or wherever) is distint from NSL; it's not a complete language.

Still, it's a great insight, and I wonder if there is a similar effect for NSL speakers of different generations (the first generation are less linguistically capable than more recent generations, who've learned the language earlier and had more interactions with a large, stable, linguistic community).
posted by lesli212 at 1:05 PM on February 10, 2011 [2 favorites]

Argle-bargle, thanks for the clarification lesli212 - I was wondering why the articles I read didn't mention NSL. I should have read more closely.
posted by filthy light thief at 1:14 PM on February 10, 2011

And the Wired article has the following paragraphs which make the distinction clear:

Homesigners in Nicaragua are famous among linguists for spontaneously creating a fully formed language when they were first brought together at a school for the deaf in the 1970s. But many homesigners stay at home, where they share a language with no one. Their “home signs” are completely made up, and lack consistent grammar and specific number words.

Over the course of three month-long trips to Nicaragua in 2006, 2007 and 2009, Spaepen gave four adult Nicaraguan homesigners a series of tests to see how they handled large numbers. They later gave the same tasks to control groups of hearing Nicaraguans who had never been to school and deaf users of American Sign Language (which does use grammar and number words) to make sure the results were not just due to illiteracy or deafness.

Razzafrazza.
posted by filthy light thief at 1:17 PM on February 10, 2011

This was, I believe, prominently featured in the RadioLab episode Words - which stands as the most incredibly mind-blowing hour of radio programming I've ever heard in my life. I seriously can't recommend that show enough - RadioLab is consistently awesome but this show in particular completely and permanently altered the way I understand thinking.
posted by Baby_Balrog at 1:19 PM on February 10, 2011 [3 favorites]

argh... teach me to read the first comment.
posted by Baby_Balrog at 1:20 PM on February 10, 2011

I still have to read the paper, but it seems to me there are (at least) two ways that being an illiterate homesigner could hamper the acquisition of large numbers:

1. Lack of language structures in the brain, which somehow also underlies number cognition
2. Poverty of stimulus -- since the sorts of situations where numbers and counting arise are often mediated linguistically, the lack of a full, rich communication system might mean much less opportunity to learn numbers

Does the paper take a position one way or the other? It would be hard to tell the difference between these two explanations -- perhaps by finding illiterate homesigners, then attempting to (somehow non-linguistically) train them to count. I believe Everett performed experiments with the Pirahã that involved trying to teach them to count, but he was able to use language, which would ruin the experiment with the homesigners.
posted by The Tensor at 1:35 PM on February 10, 2011

Unlike the Pirahã people of the Amazon (previously) who may not have the need for specificity in large numbers, the deaf in Nicaragua are surrounded by a culture that interacts in specific numbers, yet it appears they lack accuracy with numbers higher than three or four.

It's all relative. Speakers of English often use the words like "gazillion" and phrases like "boat load" when specificity in large numbers is not needed.
posted by three blind mice at 1:36 PM on February 10, 2011 [3 favorites]

replying to The Tensor - Spaepen and her coauthors definitely discount the second. They note that the homesigners they studied are fully integrated in a culture in which the non-deaf (who are also subject to the same lack of formal education) use numbers larger than 1 & 2 with precision.

But as for the first, they don't really get into deep speculation about the development of various language-related brain structures. They throw out a few theories, though. The main speculation seems to be that it may have something to do with the process of learning about numbers by counting. I wonder if the "Chinese speakers are better at math" debate may have influenced their thinking here. I might be off here, but isn't it speculated that the way numbers are spoken in chinese is more "countable" and lends itself to a better natural understanding of how numbers relate to each other?

I think their most interesting route for further investigation might be seeing how memory relates. They speculate that the home signers may not have developed enough stable symbols to overcome working memory limits about "how many" (ie, the kind of memory that focuses on what you're doing now, in contrast to long term memory).
posted by lesli212 at 1:56 PM on February 10, 2011 [1 favorite]

Compare this with the Al-Sayyid sign language in Talking Hands. It's a fascinating book that chronicles people studying the language as well as discussing the bigger picture question of whether or not sign is a language, how languages and sign languages change over time, how sign languages typically spring up and how long it takes to grow from protolanguage to full language etc, etc. There have been some studies that contrast how homesign grows into a full language and the difference in brains structures of people who are deaf with non-deaf parents, especially with regards to being able to understand point of view as well.

It is a very thorough and fascinating book. In particular, I found the studies regarding stroke-induced aphasia in the deaf completely engrossing - it is quite common, when the language center of the brain is affected, that the person loses the ability to access the signs for specific words even though he/she is capable of the gesture. For example, one patient was unable to access the sign for flower, but could only draw a spatial picture of a stem and petals.

I also found it particularly striking what an a-hole Alexander Graham Bell was towards signing. His legacy lives on at the Clarke School for the Deaf in Northampton, MA, where served on the board of trustees. The Clarke School, for better or worse, does not support sign in the school in any way. The was part of Bell's eugenicist plan to attempt to breed deafness out of the world. I worked with a man whose brothers are both profoundly deaf and went through the school. They learned what was referred to as "bathroom sign" which was the only place in the school where students could get away with signing.

I clearly have a bias. I won't say that I really know ASL, but we used Pidgin Signed English (which uses ASL signs in English word order, but without the fingerspelling cruft of Signed Exact English) with my daughter who was likely to be speech delayed due to having Down syndrome and a stroke at birth. We started with it when we believed that she was strong in language, although not in speech.

Because of my exposure, I've come to appreciate the beauty and economy (as well as the frustration) in ASL.

Recently, I found this interpretation of Jonathan Coulton's "Re: Your Brains" done by Stephen Torrence. He really does a terrific job with the facial expression aspect of ASL. If you watch this, be sure to turn on CC in the video so you can see the lyrics as well as the ASL gloss. As a side note here, my kids love this song and I know that I will pay for it, as my daughter picked up signing "eat brains" after one watching. Oh well - sometimes you accept that social price for a good developmental skill. Another side note, when he signs "I really wish you'd let us in" (PAST WISH YOU INVITE-[us]) at 0:21, the sign for wish and hunger/hungry are the same. While I don't know if it was an intentional pun, it's still pretty hilarious in the context of zombies.

And while we're at it, you might enjoy this interpretation of Cee Lo Green's "Fuck You" (NSFW, obviously). Again, I'm hardly fluent in ASL, but her facial expressions are, IMHO, not as good as Stephen Torrence, but you can really see how pronouns work in ASL - they are spatial. So unlike English, you can set in space pronouns and accurately and tersely express very complicated relationships. In this case, she sets a side a space to her left for the girl which she signs "fuck her too" (at about 0:37). In addition, at about 0:50, she doubles up "fuck you" with both hands which can be interpreted as "fuck both of you". Nice and compact. I was just thinking about how relatively easy it would be to sign "fuck you and the horse you rode in on" if you had already been talking about the horse and set it as a pronoun.

I had intended to put a lot of this into its own FPP, but this seemed like an appropriate place to threadjack.
posted by plinth at 1:58 PM on February 10, 2011 [10 favorites]

Spaepen and her coauthors definitely discount the second. They note that the homesigners they studied are fully integrated in a culture in which the non-deaf (who are also subject to the same lack of formal education) use numbers larger than 1 & 2 with precision.

I'll have to read the paper. I saw that claim mentioned in one of the articles about the research, but I'm skeptical of how "fully integrated" someone can be into a culture when they not only don't speak that culture's language, but lack any "full" language.

(I definitely need to read the paper.)
posted by The Tensor at 2:34 PM on February 10, 2011

copypasta from the paper:

The four homesigners show no congenital cogni- tive deficits and performed as well as hearing siblings and friends on tasks testing mental rotation skills. They hold jobs, make money, and interact socially with hearing friends and family.

We first asked whether homesigners were familiar enough with their society’s moneyed economy to make correct judgments about monetary values. We designed a series of tasks to assess their recognition of currency (shown, for example, by gesturing a five and two zeroes for a 500-unit bill); their ability to compare the relative value of the currency (shown, for example, by determining whether a 10-unit or a 20-unit bill has more value); and their ability to compare coins and bills of different currency type (shown, for example, by determining whether a set of nine 10-unit coins has more or less value than a 100-unit bill). All four homesigners identified money correctly and could assess its relative value, and three of the four homesigners performed with greater than chance accuracy on all monetary tasks.

BUT THEN:

Homesigners were 100% accurate on target sets of one, two, and three for both time-unlimited (Fig. 2A) and time-limited (Fig. 2B) versions of the task.† However, they performed significantly less well on target sets greater than three on both versions, coming close to the value of the target number but achieving the exactly correct value on fewer than half the trials.

Ain't it nuts? I think it just shows how amazing humans are, how much language is so part of our identities as humans; even when cut off from access to language, we still try.!
posted by lesli212 at 2:42 PM on February 10, 2011 [1 favorite]

And plinth, your daughter is SO FRICKIN CUTE! eat brains, haha.
posted by lesli212 at 2:43 PM on February 10, 2011

I'm even more suspicious now. The signers are illiterate, but can recognize "gesturing a five and two zeroes for a 500-unit bill" -- does that mean they know their numbers but not their letters? Also, they can "[determine] whether a set of nine 10-unit coins has more or less value than a 100-unit bill"? How do the authors propose they're doing that without understanding numbers above three?

(I definitely need to read the paper.)
posted by The Tensor at 2:53 PM on February 10, 2011

I'm even more suspicious now. The signers are illiterate, but can recognize "gesturing a five and two zeroes for a 500-unit bill" -- does that mean they know their numbers but not their letters?

Any American, literate or not, still learns the correspondence between the character string $20, the word "twenty" and the twenty-dollar bill. Don't see why it should be any different in Nicaragua.
posted by nebulawindphone at 3:16 PM on February 10, 2011

Okay, here's the relevant bit from the (obnoxiously non-free) PNAS article.

In the second version of the “which is more” task, participants saw 40 trials, each of which contained two sets of coins or bills. In this task, the coins on each side of the comparison were all of one kind, but homesigners had to compare across types of coins or bills. There were four types of trials, ranging in difficulty, as follows:

Type I, which included four trials, showed small numbers of bills or coins of the same denomination. These were controls to make sure homesigners were on task (e.g., two C$5 coins vs. three
C$5 coins).

Type II, which included 12 trials, showed small numbers of bills or coins of different denominations (e.g., three C$20 bills vs. one C$50 bill, two C$5 coins vs. three C$10 bills).

Type III, which included 12 trials, showed large numbers of bills or coins of different denominations with a small monetary difference between sides (e.g., six C$5 coins vs. two C$20 bills, seven C$10 bills vs. four C$20 bills).

Type IV, which included 12 trials, showed large numbers of bills or coins of different denominations with a large monetary difference between sides, always at a 1:2 or 2:3 ratio (e.g., 10 C$10 bills vs. two C$100 bills, 12 C$5 bills vs. three C$10 bills).

Trials of different types were interspersed in a quasi-random fixed order. Side of the winner, the larger number of coins being the winner, and the larger denomination being the winner were all counterbalanced across trials within each type. Moreover, within each type, homesigners saw four trials that compared coins to coins, four trials that compared bills to coins, and four trials that compared bills to bills.

Fig. 1 indicates the homesigners’ and the unschooled hearing controls’ performance on this task. Both groups did quite well. The hearing controls were far above chance on all trial types (one-sample t tests: all P < 0.01). The homesigners were above or marginally above chance on all trial types except type III [one-
sample t tests: type I, t(3) = 7.00, P < 0.01; type II, t(3) = 3.323, P < 0.05; type III, t(3) = 2.211, P = 0.114; type IV, t(3) = 2.920, P = 0.06].

There's also a graph showing how accurate the homesigners were. It looks like they could do task I (e.g. "two fives are worth less than three fives") 90-100% of the time, but task III (e.g. "six fives are worth more than two twenties") only about 70% of the time. That's better than chance, but still consistent with the idea that they're not thinking in exact numbers, just estimating — the same way I might if you asked me "are three screaming toddlers louder than a jet engine?" or "will ten watermelons fit in your trunk?" or whatever.
posted by nebulawindphone at 3:33 PM on February 10, 2011 [1 favorite]

[O]thers, usually from disadvantaged households, arrive at school not knowing what “two” means.

This made me want to go find a kid to hug (and give, like, blocks to.)
posted by SMPA at 4:01 PM on February 10, 2011

I'm even more suspicious now. The signers are illiterate, but can recognize "gesturing a five and two zeroes for a 500-unit bill" -- does that mean they know their numbers but not their letters?

It also bears remembering that "illiteracy" is a matter of degrees, and that people who are functionally illiterate or completely illiterate often employ heuristics for the tasks that are of practical urgency for them. (Money is probably just such a thing.) A few years back, I was living in California and doing adult literacy tutoring at a local library. The guy I was tutoring was in his mid 40s at the time and read at about a 2nd grade level, but he had a high school diploma and a CA driver's license. Now, I know you're thinking, "High school diploma? That's just social promotion!" But even with social promotion, most people in his position just gave up on high school and dropped out. The driver's license is the really surprising part. The CA written test is loaded with the worst sort of "Which of the following is not always true of some of stop lights in all cities?" questions you can imagine, and you need a surprisingly high score to get the license. (I have a PhD and I was sweating it a bit at the time.) But he passed it, even with a 2nd grade reading level. How? By recognizing just enough of the symbols and a small handful of words to plug into his otherwise normal cognitive processes. When I asked him how he did it, he said, "You just figure out a lot of ways to do whatever you really have to do."
posted by el_lupino at 4:16 PM on February 10, 2011 [1 favorite]

Here's another discussion where I don't belong, because while reading the FPP I was convinced there were Nicaraguan children who weren't deaf, creating their own language, and that it was a major anthropological phenomenon. And I wanted to hear more.
posted by gorgor_balabala at 3:48 AM on February 11, 2011

Ah, I see how you would think that. Again, my fault.

Reminder to self: no more posting in haste.
posted by filthy light thief at 7:20 AM on February 11, 2011

also relevant: subitizing
posted by nangar at 2:22 AM on February 12, 2011