Outline 5 - FARKAS PDF

Download as pdf or txt
Download as pdf or txt
You are on page 1of 6

Instructor: Imola-Ágnes Farkas

[email protected]
Office hours: Tuesday 9-10 a.m., M16

Introduction to Generative Grammar


Course 4:
- the aims of GG
- child language acquisition: behaviourism, the poverty of stimulus argument (Plato’s problem),
nativism, stages in child language development, features, the logical problem of child language
acquisition,
- I-language and E-language
- principles of UG (2)

Outline1

The aims of GG - summarized in a number of questions:

1. What constitutes knowledge of language?


Answer: description of grammar

2. How is such knowledge acquired?


Answer: description of the process of language acquisition

3. How is such language put to use?


Answer: description of the process of generating language

4. What are the physical mechanisms that serve as the material basis for this system of knowledge and for
the use of this knowledge?
Answer: a physical system which correlates with mental knowledge, the link between mind and brain
neurolinguistics

5. How did language evolve?

Language acquisition before GG

Behaviourism: --incorporated the structuralist tradition


--described language acquisition in terms of:
- responding to stimuli
- parroting the language of adults
- correction and reinforcement provided by meticulous care of the parents/caretakers
- connectionism

Chomsky’s critique of behaviourism introduced the key notion of creativity: people understand and produce
sentences they never heard before, making use of a finite set of means for an infinite production of sentences.
(infinite use of finite means)

1
You find in this outline the content of the slides that I project during the course, which contain the main topics and also structures
and diagrams which may be difficult and time consuming for you to copy during my lecture. They are made available to you before
class to save time and to make note-taking easier, but not unnecessary!
The outline as such (without your notes covering the detailed explanations that I give during the course) cannot constitute a sufficient
source of information when preparing for the exam. If you miss the class, it is strongly recommended that this outline be used as a
guide to the bibliography indicated at the end of this document.
Page 1
Problem 1: Acquisition through imitation
- responding to stimuli
- parroting the language of adults

Refutation:
--the first sentences children produce are simple declarative sentences, in an environment in which they are
mostly exposed to questions and commands, and only 25% declaratives
--children’s errors reflect grammatical rules, not grammar violations
--children’s knowledge of language goes beyond their linguistic input

Problem 2: Learning through reinforcement


- correction and reinforcement provided by meticulous care of the parents/caretakers

Refutation:
--language acquisition cannot be equated with learning table manners
--parents generally pay attention to the content, not form of what children say. Normally they don’t provide
negative evidence
--on the rare occasions when they do, children don’t pay attention to negative evidence

Problem 3: Language learning through association


- connectionism
--connectionist models account for some children’s overgeneralizations
and morphological acquisition

Refutation:
--although children overregularize verbs, they never overregularize auxiliaries (*I haved eaten), which
means that they operate with symbolic content in order to distinguish between auxiliaries and lexical verb
--connectionist models can generate some syntactic rules, but they cannot explain the constraints that apply
to those rules
-- poverty of stimulus problem: see pidgin and creole languages

The poverty of stimulus argument

Does the environment shape language acquisition? How rich is the linguistic environment?

Plato’s problem: the poverty of stimulus—nature vs. nurture

To what degree an organism is shaped by the environment and what is the degree to which it is the consequence
of its essential nature?
“Meno”: a slave discovers the principles of geometry without the benefit of direct instruction. Plato argues that
this kind of knowledge is grounded in the nature of reason rather than in one’s experience of the world.
Language acquisition: from a limited and rather poor linguistic evidence the child extracts the rule and then
extrapolates it producing novel utterances.

The ability of learning a language depends on a genetically determined, innate property of the mind: the
language faculty

Universal grammar
--mental generative procedure that uses finite means to generate an infinite number of sentences
--psychological entity, not a set of grammatical structures
--basic rules and a number of constraints (on form or on meaning) generate various grammatical structures
Page 2
Nativism or the innateness hypothesis
The language capacity is:
--innate
--richly structured (rules, constraints, parameterization)
--its acquisition genetically programmed and presupposes interaction of inborn factors with linguistic
environment

Effects:
-- despite limitation and learning conditions language acquisition is possible. (poverty of stimulus)
-- irrespective of the language acquired, we observe the same milestones in language acquisition. Also,
grammatical knowledge becomes progressively (in stages) visible to the child’s brain
--children’s errors observe grammatical rules
--children’s innate grammatical knowledge can lead to creolizations
--isolation from a linguistic environment during the critical period of language acquisition can lead to
failure to acquire, fully or partially, natural language

Stages in child language development

Phonological Development Development


AGE perception/ /production of /production of syntax Observations
production vocabulary and morphology
3 months crying and cooing non-communicative
4 months babble babbling is interactive and
social; infants engage in proto-
conversations with care-takers;
deaf infants engage in sign
babbling
5 months recognize their own recognize their own recognizing their own names
names; names; helps them with speech
babble; passive acquisition of segmenting and acquisition of
imitate simple vocabulary new vocabulary;
sequences
7 months can distinguish non- passive acquisition of
mother tongue vocabulary
phonemes;
babble begins to
conform to mother
tongue phonological
patterns
1 year lose the ability to content words; Holophrastic stage (one- First words (last words in
distinguish non-mother rarely use word sentences) sentences are most noticeable)
tongue phonemes underextension; En-nouns;
frequently use Korean-verbs
overextension;
understand about 70
different words;
use about six words

2 years rapid development of Two-word stage


vocabulary (Subject+verb);
no functional words or
morphemes
2 years and a half Multi-word stage
Limited use of functional
words or morphemes
(Telegraphic speech);
Beginning of the
development of
inflectional and
derivational morphology;
Use of overgeneralization

Page 3
Features of child language acquisition

-- it needs exposure to a linguistic environment

-- it occurs without explicit teaching/systematic instruction

-- occurs within a limited period of time (by the age of 5 they are proficient in their mother tongue)

-- it’s an effortless process

-- follows the same path of development, irrespective of the linguistic environment, or cultural attitudes related
to child language acquisition;

-- negative evidence is rare and has little impact.

Feral children (wild children, who grew up in the wild, or children who, due to parental neglect or abuse, were
isolated from a community of speakers) who lacked exposure to language during the critical period of language
acquisition will never be able to catch up, and will remain often mute or with poor language abilities.

The Bristol experiment 1985 (The child is 24 months old)

Child: Nobody don’t like me.


Mother: No, say “nobody likes me.”
Child: Nobody don’t like me.
Mother: No, say “nobody likes me.”
Child: Nobody don’t like me.
Mother: No, say “nobody likes me.”
Child: Nobody don’t like me.
Mother: No, say “nobody likes me.”
Child: Nobody don’t like me.
Mother: No, say “nobody likes me.”
Child: Nobody don’t like me.
Mother: No, say “nobody likes me.”
Child: Nobody don’t like me.
Mother: Now, listen carefully, say “nobody likes me.”
Child: Oh, nobody don’t likes me.”

First Language (L1) acquisition: children do not copy their parents’ speech. The errors children make reflect a
rule of grammar. In English they overgeneralize the regular plural (mouses, foots, tooths) and regular tenses (he
gived, goed, she be’s at work, I finded Renee), forms they could hardly have heard from their caretakers.
Irregular forms seem to be acquired at a later stage as exception to the rule.

The logical problem of child language acquisition

--children acquire rich linguistic knowledge that outpaces the input from the environment

--children are sensitive only to positive evidence

--children produce “errors” when they generalize from the linguistic input, or “errors” that alien to the linguistic
input, but perfectly acceptable in other languages

Page 4
--children’s “errors” would never violate basic linguistic principles, such as structure-dependency

The concept of language revisited

I-language (internalized language) = speaker’s knowledge of language, his/her linguistic competence at


different stages in its development.

E-language (externalized language) = performed speech acts in the linguistic environment the speaker is
exposed to; the use of language as a product of human behaviour.

Bristol experiment (1985):


Child’s I-language generates “Nobody don’t like me” in contrast to the child’s exposure to E-language/
environmental “Nobody likes me”

Child language acquisition can be described as a process in which I-language (universal innate linguistic
competence) gradually narrows down to the specificity of E-language (linguistic environment)

Redefining theoretical approaches to the study of language

I-language (internalized language) approach: describes/ researches the speaker’s knowledge of language,
his/her linguistic competence; a grammar that describes the properties of the human mind. It is concerned with
what a speaker knows about language and where it comes from.

E-language (externalized language) approach: language produced in concrete situations, it focuses on


performance, concentrates on language in a social context and associates it with human behaviour, not with an
internal psychological state, or the properties of the human mind

Principles of UG

A. The embedding principle: (clause recursion)

John loves Mary.

Bill suspects that John loves Mary.

George knows that Bill suspects that John loves Mary.

Sue believes that George knows that Bill suspects that John loves Mary.

B. The structure-dependency principle

knowledge of language relies on the structural relationship in the sentence rather than on the sequence of words.
A sentence like The child drew an elephant is not a string of words in a sequence

sentence

the + child + drew + an + elephant

Page 5
but structured construction which can be broken up into smaller constituent which may be as well structured
constructions…

Example of the validity of the structure-dependency principle:

Fronting the auxiliary in English interrogatives:


Which is the rule? Move the auxiliary to the front! ????

[[The man [who is1 tall]] is2 John.]]

*Is1 the man who tall is2 John?

[Is2 [the man [who is1 tall]] John?]]

“Children unerringly use computationally complex structure-dependent rules rather than computationally
simple rules that involve only the predicate ‘leftmost’ in a linear sequence of words” (Chomsky).

UG theory claims that such principles are inherently impossible to learn; if they are not learnt, they must be
part of the human mind, that is innate.

Bibliography:
Guasti, Maria Teresa. (2002). Language Acquisition. The Growth of Grammar. MIT press, Cambridge, Mass.
pp. 1-20.

Vivian Cook & Mark Newson. (1996). Chomsky’s Universal Grammar. Second edition, Blackwell, pp. 1-39.

Page 6

You might also like