Talk:Language acquisition

From Wikipedia, the free encyclopedia

Contents

[edit] empty elements

It is known from early work on phrase structure grammar that grammars with empty elements can be converted into equivalent grammars without empty elements. Yehoshua Bar-Hillel and M. Perles and E. Shamir, 1961, On Formal Properties of Simple Phrase-Structure Grammars. Zeitschrift für Phonetik, Sprachwissenschaft und Kommunikationsforschung. 14 (2):143-172. However grammars without empty elements cannot express things linguists want to express in a satisfying way. In the introductionary textbook on HPSG by Sag, Wasow, Bender [1] an empty head is suggested for the analysis of copulaless sentences in African American Vernacular English. Emily Bender argued at length in her dissertation that the empty copula is the only way to capture the facts in a non-stipulative manner. You may also read my paper about empty elements and surface-based syntax at [2]. In any case the claim that HPSG does not use empty elements is wrong. The analogon to movement is a mechanism that handles unbounded dependencies. I guess every theory needs something like that.

Stefan Müller, November, 5th, 2005

[edit] References missing, needed, or incomplete

  • In there are significant studies in biogenetics there's no reference/source mentioned.

[edit] stages of acquistion?

This article could really use some content on the stages children go through in L1 acq. (babbling, 1-word, 2-word, telegraphic, full language). Roy Allahades 06:10, 12 Oct 2005


[edit] rename the article ?

L1 and L2 acquisition are related however-- just look at the morpheme acquisition studies that have been done for English as both an L1 and an L2.

http://www.google.com/search?client=safari&rls=en&q=acquisition+orders+in+l2&ie=UTF-8&oe=UTF-8

Instead of breaking this into two articles, we need to add a section discussing differences in L2 acquisition (and linking to the stub on Interlanguage. Roy Allahades 06:00, 12 Oct 2005


In my experience, this field has been and is called "Child's Acquisition of Language" to distinguish it from, say, adults learning a second language.

There are 3 very closely-related ideas here: "learning a second language", "initial acquisition of language by children from adults", and "how did all these languages get started ?" (Origin of language). They are all kinds of language acquisition. How do we avoid repeating much of the same stuff in 3 different articles ? --DavidCary 02:19, 27 Oct 2004 (UTC)

Should language development be separate from this? If not, should this be renamed to cover both the acquisition and the subsequent development, and also the development of languages other than the first? Angela. 05:35, 30 Dec 2003 (UTC)

(has this already been merged ?)

When I have studied linguistics, the term that I've encountered most often is first language acquisition (to make the difference between it and second language acquisition).

[edit] other discussion

I'm terribly sorry if I've made a good deal of errors with this entry! I'm new to Wikipedia and still trying to get a better idea of what needs to be done.-Devotchka


The most popular opinion is that language acquisition results from both nurture and nature. This theory was introduced by Noam Chomsky.

The two sentences don't work for me. The first doesn't really say anything, and the second is too broad, I think. I'll try to reword them better; please check that I don't totally change what you were trying to say.---- Then welcome and relax. You are only making the same kind of mistakes almost all of us have as beginners. RoseParks

just a thought: nature and nurture are the two opposing views. nature (i.e. the innateness hypothesis) was proposed by Chomsky, while the nurture approach (i.e. language is acquired by general learning mechanisms) was introduced by people like Piaget. so saying that "this theory" was introduced by Chomsky is a bit misleading, since he really is a defender of the nature approach only. AK

[edit] Argument from the Poverty of Stimulus

The characterisation of the Argument from the Poverty of Stimulus is just plain wrong. Chomsky and others have made it clear that it's not about *degraded* input (incomplete or ungrammatical utterances) but about grammatical phenomena that cannot in principle be learned on the basis of positive data alone. To reduce it to degraded input sets up a straw man and misleads people. This needs to be thought through. I recommend the article on the APS in Kasher's volume _The Chomskyan Turn_ as a very good explanation of the argument and its role in Chomsky's view of first language acquisition.

The poverty of stimulus however is not a fact and has not been proven. Also, there is in fact a wealth of negative data in the speech environment, a few examples:
  • Self-corrections are stunningly common, see conversation analysis.
  • The hearers failure to understand what the child must be assuming to be a well-formed output is another form of negative data.
--AkselGerner (talk) 21:16, 29 February 2008 (UTC)

[edit] what is aquisition

[edit] etc.

Statistical methods for studies on first language acquisition http://www.biostat.sdu.dk/~wv/firstlanguage.html

nine stages of Second Language Learning http://www.bankstreet.edu/literacyguide/ellstages.html

Assisting English as a Second Language (E.S.L.) Students http://www.sasked.gov.sk.ca/docs/kindergarten/kindlang.html

"Thinking before you speak" "Researchers have found that five-month-old babies can comprehend concepts for which they have not yet learned words, thus answering the age-old question: Which comes first, an idea or the language to express it?" http://www.mirabilis.ca/archives/001977.html http://www.thestar.com/NASApp/cs/ContentServer?pagename=thestar/Layout/Article_Type1&c=Article&cid=1091139009712&call_pageid=968867505381&col=969048872038


"Dogs Similar to Children When Learning Language." http://www.mirabilis.ca/archives/001812.html http://www.washingtonpost.com/wp-dyn/articles/A31130-2004Jun10.html

Language acquisition How did you learn to speak your native language? http://www.facstaff.bucknell.edu/rbeard/acquisition.html


Proposed Experiment in Concretized Learning by Win Wenger http://winwenger.com/part51.htm

"First Language Acquisition" http://pandora.cii.wwu.edu/vajda/ling201/test4materials/ChildLangAcquisition.htm

[edit] Huge Wall of Text

This article needs to be broken up into sections TheCoffee 13:11, 29 Jan 2005 (UTC)

[edit] Clarification required I think

I have some problems with the following paragraph

However, a powerful argument against Chomskian views of language acquisistion lies in Chomskian theory itself. The theory has several hypothetical constructs, such as movement, empty categories, complex underlying structures, and strict binary branching, that cannot possibly be acquired from any amount of input. Since the theory is, in essence, unlearnably complex, then it must be innate. A different theory of language, however, may yield different conclusions. While all theories of language acquisition posit some degree of innateness, a less convoluted theory might involve less innate structure and more learning. Under such a theory of grammar, the input, combined with both general and language-specific learning capacities, might be sufficient for acquisition.

As written it doesn't explain much. If the theory of language (grammar etc) that Chomsky presents is not learnable on the input alone, then this isn't at all a powerful argument against his theory of language acquisition: it's rather the opposite, if that is his theory of language is true. Does the contributor mean that if there is a "wholemeal" theory of grammar (without inaudibilia etc) that this could be learned on the input alone without the appeal to any language specific mental devices? That is plausibly true. If for example the binding theory is not necessary to explain the distributional constraints on referring expressions and corerential expressions then the binding theory loses any power it has to motivate the assumption of an innate language faculty.

I can see that this would be worth stating in the article somewhere (though the article is in danger of becoming Chomsky heavy) but it doesn't seem to be the purpose of this contributer's paragraph since s/he writes

"nder such a theory of grammar, the input, combined with both general and language-specific learning capacities, might be sufficient for acquisition."

In which case all you've got is a different theory of the innate language acquisition device - you don't have a theory that doesn't have such an innate device and your theory is still in the most important ways Chomskyan.

I'm loath to edit this unless there is some discussion first because I am a Chomskyan and I don't want to let my committment to generative grammar colour my editing.

Jim

Someone rewrite this without the blatant Chomsky ass-kissing, please? I don't need to be told which men are made of straw, and consider it a mark of intellectual failure that the authors of this article have included and subsequently not removed such doggerel.

Once again Wikipedia proves that All Of Us is dumber than Some Of Us.

Thanks for your comment. If you feel a change is needed, feel free to make it yourself! Wikipedia is a wiki, so anyone (yourself included) can edit any article by following the Edit this page link. You don't even need to log in, although there are several reasons why you might want to. Wikipedia convention is to be bold and not be afraid of making mistakes. If you're not sure how editing works, have a look at How to edit a page, or try out the Sandbox to test your editing skills. New contributors are always welcome. — mark ✎ 08:00, 31 May 2005 (UTC)


Agreed, non-Chomskyian views definitely deserve a lot more airtime here! I did a tiny bit of NPOV but it still needs way more.

-- Gabe

[edit] language "development" in adults?

is second language acquisition or learning by adults also referred to as "language development"?, as implied by the opening lines of the article .

 Doldrums 12:02, 16 October 2005 (UTC)

[edit] broca's area

Shouldn't this article discuss Broca's area of the brain? -Jose


In what way is Broca's area relevant to the characterisation of language acquisition? Is it any more or less relevant than Wernicke's area? I don't see any reason a discussion of Broca's should be included.


It appears that Broka's area takes care of 'verbs', both in action as also in speech, whereas Wernick's area takes care of nouns. see for eg. Neural Modeling of Language Acquisition,


I have to disagree with that previous sentence. It's just plain false. --Drmarc 03:30, 2 August 2006 (UTC)


Lately there has been a marked interest on investigative and clinical studies on human brains to identify regions where different word types, nouns, verbs etc., are processed via neuropsychological or neuro-imaging techniques [1,2,3]. Apart from revealing regions of functional localizations such studies also enable identification of lesions or other damaged brain locations. It has been observed that left anterior and mid temporal lobes in human brains, Wernicke's sensory areas are associated mainly with declarative memory and operated on nouns and attended to semantic processing of the language; whereas left frontal motor regions, Broca's motor areas are associated mainly with procedural memory operating on verbs and assisting in syntactic processing of language. Any lesions or other damages in these two regions led to characteristic problems in linguistic reproduction termed respectively as Wernicke's Sensory aphasia and Broca’s agrammatical aphasia. The process by which such anatomical discriminations are developed has however remained obscure

References:

[1]. Cappa S F, Sandrini M, Rossini P M, Sosta K and Miniussi C. (2000)  : The role of left frontal lobe in action naming. Neurology, 59, pp 720-723. [2]. Damas A R and Tranel D. (1993): Nouns and Verbs are retrieved with Differently Distributed Neural Systems. Proc. National Academy of Sciences, USA, 11, pp 4957-4960. [3]. Federmeier K D, Segal J B, Lombrozo T and Kutas M.(2000) : Brain responses to nouns, verbs and class-ambiguous words in context. Brain, 123, 12, pp 2552-2562.

==== PPRao. Aug 15, 2006.

[edit] Different theories

Should the article include different (nativist) theories of language acquisition? e.g. the cue-based model? Or do we already have articles for those? - FrancisTyers 20:25, 4 December 2005 (UTC)


[edit] Baker

Baker's work is very controversial, however, because he has argued (1996: 496-515) that principles and parameters do not have biological or sociological origins, but instead were created by God (i.e. creationism).

I removed this pending a better reference. I've never heard anyone working in p&p ever argue that "principles and parameters were created by a god". - FrancisTyers 18:54, 13 December 2005 (UTC)

The reference is at the bottom of the page.

Yeah, but I don't have that book, I have The Atoms of Language and Baker doesn't mention god or creation once in that, which would be unusual since he is talking about the same thing. - FrancisTyers 19:19, 13 December 2005 (UTC)

Yes, it is unusual, but true, as you will see when you read the book.

Could you type out the paragraph where he refers to either god or creationism? Thanks. PS. You can sign your post using ~~~~ - FrancisTyers 19:41, 13 December 2005 (UTC)

Here are some excerpts from Baker's 4 page discussion of his "theological" explanation for the origin of principles and parameters:

512 “If biological and sociological categories are ill-equipped to explain the possibility of syntactic diversity, are there are other candidates?…Since God is the quintessential spiritual being, and language has its origins in him, language is therefore a spiritual phenomenon…In other words, humanity is given a spiritual nature [by God] that is specifically said to be parallel in many respects to God’s. Among other things, this means that since God is a linguistic being, so are humans.”


512-3 “Strikingly, the problem of syntactic diversity is one aspect of language for which the Judeo-Christian scriptures provide an explanation…[A discussion of the Tower of Babel]...[God] creates linguistic diversity to make difficult or impossible large-scale and long-term cooperative efforts. The reason for this (given that God is good) is presumably that such endeavors have an inherent tendency toward evil and destructive manipulation”

BFSkinner

Thanks for the quotes, my university has the book in the library so I'll check it up tommorow. From what it sounds like the chapter on a theological explanation is a rhetorical look at the options and not a statement that he believes in creationism, or that god created language. Feel free to add it back in, but I'll verify it later. PS. I hope you understand my suspicion, especially considering your username ;) - FrancisTyers 20:14, 13 December 2005 (UTC)

The following additional quotations indicate that Baker does, in fact, believe in creationism. Of course, people should read the original chapter to decide for themselves.

513 "Suppose that these [spiritual] traditions are either more insightful or have better memories, or both, than the dominant modern academic tradition in this respect."

514 "However the historical details of the story [of the Tower of Babel in Genesis] are to be taken, its basic point is clear: linguistic diversity results from a direct act of God. This act was logically distinct from the act that gave humanity a linguistic nature in the first place. If macroparameters are one of the important mechanisms giving rise to serious linguistic diversity (diversity that cannot be overcome by ordinary lexical learning), then their origin is distinctly spiritual in nature."

515 "In closing, I should say that I do not intend these last few pages to single-handedly convince those who are materialists in theory or practice that there are spiritual forces at work in the world. However, it does seem right that those who are already convinced of this be alert to places where spiritual forces may shed some light on important facts of intellectual interest...I suggest that part of the meaning of the Polysynthesis Parameter is that we are not simply biochemical survival machines or the cells of a social organism; rather, we are spiritual beings that in part transcend those forces."

BFSkinner

Fair cop on Baker. I just looked it up and it seems he is a creationist. Who'd have thought it eh ;) He's probably toned it down in more recent work so he doesn't take so much flak. - FrancisTyers 21:08, 14 December 2005 (UTC)

My2Cents

I think that to what extent Baker views parameters as coming from God, Darwin, or anywhere else is immaterial in the context of a discussion of first language acquisition. While the models of the biological evolution of "parameters" can seem fanciful (there are others that are less so), the more troubling problem, and the more important one for THIS topic, is that as appealing as parameters are for typological classification and generative syntax theories, research in acquisition has had trouble showing that this is actually what children are doing when they acquire language. However, they remain an important theory since they provide a model of adult syntax that is, theoretically, learnable. It is important to bear in mind that the theoretical underpinnings of parameters as an acquisition theory are linguistic, not evolutionary or creationistic. Therefore, Baker's (or, for that matter, Chomsky's) speculations on the subject are really premature until the "fact" of parameters in adult syntax is resolved with an account of what children are actually doing when they acquire them.

[edit] External links

Its that time again, external links pruning. See Wikipedia:External links. - FrancisTyers 00:15, 19 January 2006 (UTC)

[edit] Genie

"An interesting example of this is the case of Genie, also known as "The Wild Child". A thirteen-year-old victim of lifelong child abuse, Genie was discovered in her home on November 4th, 1970, strapped to a potty chair and wearing diapers." - Genie was not discovered in her home. The abuse came to light when she accompanied her mother to a government building as her mother sought disability allowences.

Sign your comments. If there's a problem with the article, feel free to fix it. That's the whole point of a Wiki. --LakeHMM 03:09, 2 May 2006 (UTC)

[edit] Average Age

Can someone please add the average age a child learns to speak? Not a huge range - *average*. --LakeHMM 03:10, 2 May 2006 (UTC)

The poster raises a really good point here: there's nothing here that covers the basic facts about language acquisition, having to do with developmental milestones. Really, what we have right now is a really nice summary of theories, but it's worthwhile including something for folks who are not so keen on the linguistics/cognitive science end of things. --Drmarc 03:34, 2 August 2006 (UTC)

[edit] DeKeyser

Regarding "DeKeyser talks about the role of language aptitude as opposed to the critical period," should this be removed? From what I've seen DeKeyser talks about both, as they are two separate ideas.

[edit] Reply to querry on Neanderthal Man .. .. under "Neuro-Modeling .. .. "

For the query on Neanderthal man, my thanks for the interest and I offer following comments: i) It is contended that reasons for all four events, viz, extinction of neanderthal man, emergence of a large cortex, of modern homo-sapien and of language skills are concurrent, guided by ‘Occams’ razor’; ii) This issue is not in main line of the theme here, ie., neural modeling, and iii) I’ve begun to doubt if the messages in this section as also in related sections under ‘Syntax’, are getting communicated to the readers or not for want of elaborations, clarity or lack of conventionality etc. Any way, a couple of sub-sections on ‘Sentences’ and ‘A few remarks on ‘Verb Software’ are planned before conclusions.

[edit] Question re. validity of reference to Neanderthal

About 100 thousand years ago, the Neanderthal man with his shallow forehead got on to the track of speech and auditory skills and evolved into modern Homo sapiens with a raised forehead and cranial structure [citation needed].

— snip —

I think this progression as written may be out of date. It is my understanding that both Neaderthal man and modern man -- Homo sapiens -- evolved from the same earlier ancestor in two different branches. That is, Homo sapiens did not evolve 'from' Neanderthal man but 'parallel to' Neanderthal man. See reference below.

http://www.cnn.com/2006/TECH/science/07/20/neanderthal.genome.ap/index.html

"At the Max Planck Institute, the project also involves Svante Paabo, who nine years ago participated in a pioneering, though smaller-scale, DNA test on a Neanderthal sample.

That study suggested that Neanderthals and humans split from a common ancestor a half-million years ago and backed the theory that Neanderthals were an evolutionary dead end, not a direct ancestor of modern humans.

The new project will help in understanding how characteristics unique to humans evolved and "will also identify those genetic changes that enabled modern humans to leave Africa and rapidly spread around the world," Paabo said in a statement Thursday."

Prabhakar P Rao 11:27, 11 August 2006 (UTC)

[edit] Neanderthal Man & Firming-up-winning-Synapses

The statement in the text on the Neanderthal man implying multiple percepts is hoped to be clarified by the Neanderthal genome sequencing project being jointly pursued at Max Plank Institute in Germany and 454-Life Sciences Corporation at Connecticut. Now that a question mark, (citation required), has been superscripted on the statement, it should suffice to retain it so for the moment and we may proceed on other avenues. For example, consider the ‘firming-up-winning-synapses’ learning process:

A ‘firming-up-winning-synapses’ learning (somewhat akin to artificial neural network learning) process has been quoted in ‘Bio-neural Software model for Verbs’. A question might arise as to what happens to the ‘non-winning synapses, do they wither out?’ In general biological contraptions are not much wasteful. It is possible that the firmed-up synapses are soon delineated out for the set service and the remnant neuron pool will proceed on a hunt for finer details on information [1]; ie, in initial iterations, toddlers recognize generic animals; on a second round they use remnant neuron pool to recognize cats, tigers, birds, fish etc; they proceed further on finer classifications, such as wild, domestic, flying etc, and finally on to recognize individualities too, such as cows yield milk, dogs run after cats that run after rats, parrots talk etc. Perhaps some artificial neural circuit configurations can be devised in this direction?

Reference: [1]: http://arvo.ifi.unizh.ch/~andel/neurowiki/nw.cgi/SynapticPlasiticity/

=== PPRao, July 23, 2006.

[edit] Concurrent Model for Nouns .. ..

[edit] Concurrent Model for Abstract Nouns?

Development of concurrent neural networks for representation, ie., of objective nouns, such as teddy, sweet cake etc, and their meanings in the associational cortex by toddlers as part of their first language acquisition has been brought out in the text[1]. How about abstract nouns, (of course not for toddlers, but say, for adolescents) such as, battle, empathy, values etc? Biological systems often use any facility for multiple services (eg, our blood circulation system) as also, they reuse any methodology acquired for similar other services. It might thus be worthwhile to contemplate on how such abstract nouns get represented in our brains/minds and what kind of concurrent neural networks could enable their recognition, perception.


[edit] Concurrent or Connectionist Model?

On some reflections it appears that the ‘Concurrent Model for Nouns’ introduced here has many features paralleling the “Connectionist models, which emphasize the idea that a person's lexicon and his thoughts operate on some kind of (neural) network”[2] and the ‘Bio-neural Software Model’ introduced here appears to possess a few features of the “Nativist models, which assert that there are specialized devices in the brain dedicated to language acquisition” [3] with the implication that these devices refer to the Bio-Neural Software acquired by the toddlers. In order to avoid unnecessary duplicities, I propose to place ‘Connectionist Model’ in place of ‘Concurrent Model’ above. The other term, ‘Bio-neural Software Model for Verbs’ has however, certain specific functionality implied; hence it shall remain as it is.

References 1.Concurrent Model for Nouns & their Meanings :http://en.wikipedia.org/wiki/Language_ Acquisition

2.Connectionism: http://en.wikipedia.org/wiki/Connectionism/

3 Psychological Nativism: http://en.wikipedia.org/wiki/Psychological_nativism/

=== PPRao, Aug 11, 2006.

[edit] Too much theory, not enough data

This article covers alot of the theories of how children learn languages, but it doesn't give any data on when they start speaking or what language skills they acquire in what order. What is the average age at which children start pronouncing recognizable words? When do they start forming sentences? At what average age have they completely learned the phonology of their language?

Well the field is already all theory... Mdoff 20:26, 7 January 2007 (UTC)

I am surprised no mention has been made in this article of the work of Fiona Cowie:

Cowie, F. (1997), 'The Logical Problem Of Language Acquisition', Synthèse 111, pp.17-51.

Cowie, F. (2002), What's Within. Nativism Reconsidered (Oxford University Press).

Or that of Geoffrey Sampson:

Sampson, G: The 'Language Instinct' Debate (Continuum, 2004, ISBN 0-8264-7385).

Rosa Lichtenstein 07:02, 13 July 2007 (UTC)

[edit] Huh?

I was never able to understand this part of this paragraph:

"However, an argument against Chomskian views of language acquisition lies in Chomskian theory itself. The theory has several hypothetical constructs, such as movement, empty categories, complex underlying structures, and strict binary branching, that cannot possibly be acquired from any amount of input. Since the theory is, in essence, unlearnably complex, then it must be innate. A different theory of language, however, may yield different conclusions."

Perhaps it can be reworded? 66.21.203.42 19:12, 15 July 2007 (UTC)

[edit] War of Arguments

NPOV needed in the Criticism and alternative theories section, please. this is not about arguments, but about language acquisition. the one who wrote this section starts saying "(...) Nevertheless, Snow's criticisms might be powerful against Chomsky's argument, if the argument from the poverty of stimulus were indeed an argument about degenerate stimulus, but it is not. The argument from the poverty of stimulus is that there are principles of grammar that cannot be learned on the basis of positive input alone, however complete and grammatical that evidence is. This argument is not vulnerable to objection based on evidence from interaction studies such as Snow's.

However, an argument against Chomskian views of language acquisition lies in Chomskian theory itself." —Preceding unsigned comment added by 189.156.27.184 (talk) 03:37, 7 January 2008 (UTC)

[edit] Skewed article

The bias in favor of generativist theory in this article is astounding. I propose that the whole thing be reorganized into a chronological account. The study of language acquisition is older than Chomsky and the article should not be focused on generativist theory. The generative and functional theories should be both given a quota (say 400 words) to avoid too much superfluous information that goes in too much detail into the theories. This is wikipedia, not a textbook. The whole critical point should be moved to the nativist section. Examples as given are not necessary and are not wikipedia material.--AkselGerner (talk) 21:26, 29 February 2008 (UTC)

[edit] Misleading wording: language-specific

"One hotly debated issue is whether the biological contribution includes language-specific capacities, often described as universal grammar. For fifty years, linguists Noam Chomsky and the late Eric Lenneberg argued for the hypothesis that children have innate, language-specific abilities that facilitate and constrain language learning[1]."

Language-specific implies that the biology includes capacities related to the specific first language of the language learner; this is NOT what Chomsky has said. Chomsky has said that the Universal Grammar has parameters that are set on the basis of first-language data, UG is NOT language-specific, on the contrary it determinalistically delimits the possible grammar of ANY language: Chomsky posits that no grammar is possible that does not fit in the binary parametres of UG. This is obviously preposterous, but not as obviously preposterous as having elements of, say, the thai language coded in the genetics of the Thai peoples, this would lead to the question of linguistic cross-breeding and other monstrosities. Chomsky may be a misguided old crank, but he's not stupid.--AkselGerner (talk) 21:31, 3 March 2008 (UTC)

Went ahead and changed it... took me a while to figure out what was probably intended.--AkselGerner (talk) 20:32, 8 March 2008 (UTC)

[edit] Check my anti-vandalism

Someone had gotten bored a while back... the vandalism had been partially corrected, but not by undoing, so I had to figure out which particular changes had already been undone and what was left. I may have missed something, so please give this article a careful read. Ken (talk) 00:28, 26 April 2008 (UTC)