[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [xmca] Deb Roy: The birth of a word

Dear Lois:
Well, the short answer is that "we" means me and my folks in my neck of the woods, which is applied linguistics, EFL, TESOL, teaching foreign languages and so on. We are a technology, rather than a science. We usually don't distinguish between learning and development, and we like stuff that's linear. We think of first language acquisition nostalgically, as a kind of Edenic time when langauge learning was painless and unconscious and effort-free.
Take, for example, the wonderful work you cite from Lois Bloom. One of the early collaborators on this work was (I think) Patsy Lightbown. Lightbown believes that vocabulary size is a "placement indicator". In this belief, she's really carrying on quite directly from the old Roger Brown school of MLU and vocabulary size that you and Lois Bloom were criticizing.
Lately, she's been working with Paul Meara on the "yes/no" vocabulary test. You get a word on a computer screen. The computer asks you if you know the word. You say "yes" or "no", and the computer, clever Dick, uses artificial words to calculate the number of times you lied and to adjust your score accordingly. Context? Never heard of it.
(If you think back to our recent discussion of Lantolf and Poehner, you'll see that my MAIN criticism of their work was that the idea that development is simply the internalization of knowledge that used to be extra-mental or inter-mental suggests a weak distinction between learning and development--one that is impervious to the CONTENT of what is being internalized and one that is gradualistic and linear.)
But the long answer, I think, is that Deb Roy HIMSELF subscribes to these three assumptions: no distinction between learning and development, more or less unconscious learning, and a fundamental, qualitative difference with second language learning in favor of the latter. 
That is why he:
 a) presents the data as a set of sound-bites (literally!) strung together in a single, seamless evolution (without initiates), 
b) does not stop and explain CLEAR crisis points (e.g. 5.04, 5.10, 5.15, 5.19, where the child strays briefly into an adult like vowel sound, and then 5.25 and 5.33, where the child sounds like his is deliberately enunciating), and 
c) sees this work as being related to work in adult communications, possibly artificial speech technology, but not second language learning.
I am not entirely sure I agree with Martin's and Jim's criticisms. First of all, when I read Halliday's work on early language acquisition, it seems MORE objective than Deb Roy's "space time worms". Halliday is looking at grammar and especially at function. But I am really not sure at all what Deb Roy is looking at. I can't even understand, when I am looking at the worms, what is space and what is time, but above all I can't understand how it helps him organize his transcriptions. (I can see how it makes for a cool presentation, though!)
I also don't think that the data can or should show the child's understanding of the concept of 'water". I suppose that understanding to be preconceptual, that is, purely functional. The child wants water, and so the child is asking for it. The child wants an adult to notice some water and so is remarking upon it. The child is wet and so the child is indicating it. I guess Jim is right to suggest that there is no concept there, but I don't really see how there could be.
What I think IS there, and what I think IS astonishing in Deb Roy's data, is the child's functional mastery of the SYLLABLE as a unit of speech analysis. I don't think "gaa" or even "gaa-gaa" suggests the unit of the syllable, because we can't really tell if the child thinks "gaa-gaa" is one word or two.
But the use of "Wa-ter" suggests:
a) The child knows what a syllable is (because the child has divided the speech sound stream into two syllables).
b) The child knows that two syllables often occur together to make a single word.
c) The child conjectures something about the possible structures of a syllable (because the word is divided at the end of a vowel, and not in the middle, so the child conjectures that CV is a possible syllable and CVC is a possible syllable, but CA-AVAC is not)
Now, of course, none of this knowledge is declarative, and all of it is what Halliday would call  functional. The child does this before the child really knows what he is doing. But that is probably the ONLY valid point of comparison with a machine.
David Kellogg
Seoul National University of Education
ue, 3/15/11, Lois Holzman <lholzman@eastsideinstitute.org> wrote:

From: Lois Holzman <lholzman@eastsideinstitute.org>
Subject: Re: [xmca] Deb Roy: The birth of a word
To: "eXtended Mind, Culture, Activity" <xmca@weber.ucsd.edu>
Date: Tuesday, March 15, 2011, 7:33 PM

I watched the Roy video a few days before it began to be discussed here and found it to be one of those TED talks that dazzles. I'm fascinated that not only a family but a funding source and research lab all put significant resources into it—and so, for me, what's fascinating is this is but one instance of what human beings get on board for and are passionate about. We're amazing in this regard!
But I failed to be impressed by the substance of the talk and thought I would share that here. The approach to understanding language development/acquisition that apparently prompted the "study" seems to me to be very outdated, not even something Lois Bloom and I might have done 35 years ago had the technology been invented and available.  For even then, we and others researching early language had moved beyond utterances in isolation to the unity,  semantics/syntax/pragmatics in/and the cultural/conversational context. 
Further, our approach (as was that of many others at the time) was looking at what we assumed was development or, to give us the benefit of the doubt (since we were not yet influenced much by Vygotsky) the unity learning/development.

So, David, when you say "we assume that first language acquisition is learning rather than development" I am truly puzzled. Who is the "we"?
Ditto the other two assumptions.
I haven't kept up with the language acquisition literature, so maybe it's the case that these assumptions are out there. But what about the distributed language literature?

Does anyone else here think that words are not born?


On Mar 15, 2011, at 9:46 PM, David Kellogg wrote:

> Now, I want to argue that this disproves, or at least severely undermines, some of our heartfelt beliefs about acquiring a first language. 
> a) We assume that first language acquisition is learning rather than development. By this we assume that it can be plotted linearly and quantitatively by measures such as vocabulary size and MLU. But the data suggests that what is going on is not the learning of a single word but rather the development of a system. What the child learns is not the single fact of the word "water", but (like Helen Keller before him) the general principle of language, in this case, the principle that words tend to divide into syllables, and that these syllables are marked by consonants and vowels. This allows the child to completely reorganize as syllables and perhaps even as consonant-vowel strings all material learned in the past, and anticipate the structure of all the material to be learnt in the future.
> b) We assume that the first language acquisition is nondeliberate, unintentional, and to a very large degree unconscious. By this we assume what we remember: we woke up one day and discovered that we could miraculously understand what people were saying to us and that the world made sense for the first time. But the data suggest that what is going on is deliberate, intentional, and if we refer to it as unconscious we mean "unconscious" only in the very special sense that the child, lacking language, does not yet have consciousness as we know it.
> c) We assume that first language acquisition is a completely DIFFERENT experimence from learning a second language in a classroom. By this we assume an invidious comparison: the first is painless and completely successful, while the latter is a daily torment with very doubtful prospects. But the data suggest that what is going on is actually quite similar to the data I will give below, and from the child's point of view, the prospects of success must seem similarly precarious.

Don't forget to check out the latest at http://loisholzman.org

Lois Holzman, Ph.D.
Director, East Side Institute for Group and Short Term Psychotherapy
920 Broadway, 14th floor
New York NY 10010
Chair, Global Outreach for UX (www.allstars.org/ux)
tel. 212.941.8906 ext. 324
fax 718.797.3966


xmca mailing list

xmca mailing list