Transcript language

Language and its Development
Language

What defines language? What properties does
language have?

How does language allow for communicative power?

Why does language exist? What does it gain us?

When does language develop?
Properties of Language

Symbolic
Cat =



Thor =
Arbitrary
Productive
Discrete

Language is based on discrete units; each unit is unique
Properties of Language
/bog/
/pog/
Properties of Language
Properties of Language

Symbolic
Cat =


Arbitrary
Discrete


Thor =
Language is based on discrete units; each unit is unique
Productive
Properties of Language- Productivity

We can say sentences we’ve never heard before



“I hate you, Mommy!”
We have a limited set of words and structures that can be
recombined.
Generativity:

“He said that she told them that he thought that we heard that
they reported that…”
Ambiguity in Speech
Resolving Ambiguity
Social agreement, context, intention
Grice (1975): Maxims of Conversation 
1.
2.
3.
4.
Quality: Tell the truth! (Avoid falsehoods *and* statements for
which you have no evidence)
Quantity: Include what is necessary to express information, and
nothing extraneous.
Utterances will be related to the topic at hand
Manner: Avoid ambiguity, use common ground (Clark)
Ambiguity in Speech

Humor:


Last night I shot an elephant in my pyjamas. What he was doing
in my pyjamas, I’ll never know”- Groucho Marx
Garden Path Sentences


The horse raced past the barn fell.
The prime number few.
Properties of Language

Is language unique to humans? Or does animal
communication meet these criteria?

Symbolic?


Arbitrary?


For some animals,YES; others, NO
Discrete


call systems used by animals only in presence of predators; NO
NO
Productive

Productivity dependent upon discreteness; NO
Language and Thought


What does language allow us in terms of our ability to
think?
Whorfian hypothesis


Eskimos and “snow”
Interlanguage differences



Colors
Spatial position/direction
Grammatical gender
The Innateness of Language

Behaviorism: Language is learned like everything else



We say something, we receive feedback, which encourages us
to say it again
BUT: We can say things we’ve never heard; we can
produce new structures.
Chomsky: Language is innate to humans



Language Acquisition Device (LAD)
Universal Grammar
Poverty of the Stimulus
The Nature of Feedback (Poverty of the
Stimulus)
1.
2.
3.
4.
5.
Children get little or no direct instruction.
Children get little feedback and don’t listen to what they get -- so
why do they ever correct their errors?
Children hear many ungrammatical structures not identified as
such -- how do they come to learn these are wrong?
In some cultures adults don’t speak to children.
Children will make up a language if they are not given one -- deaf
children of hearing parents.
Feedback
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.
[dialogue repeated eight times]
Mother:
Now listen carefully, say “Nobody
likes me.”
Child:
Oh! Nobody don’t likeS me.
The Nature of Feedback (Poverty of the
Stimulus)
1.
2.
3.
4.
5.
Children get little or no direct instruction.
Children get little feedback and don’t listen to what they get -- so
why do they ever correct their errors?
Children hear many ungrammatical structures not identified as
such -- how do they come to learn these are wrong?
In some cultures adults don’t speak to children.
Children will make up a language if they are not given one -- deaf
children of hearing parents.
The Language Gene


SLI: Specific Language Impairment: Language is impaired
without signs of impairment in other areas (motor,
cognitive, etc.)
The FOXP2 gene


Members of the KE family with a corruption of this gene had
SLI; the others didn’t.
The Language Gene?
The Language Gene
The Language Gene

Is it really specific to humans?


Is language the only area affected by the gene?


NO
NO
FOXP2 affects rapid-motor sequencing (which is
imperative for language), but not language itself.
Statistical Word Learning


Saffran, Aslin, and Newport (1996)
8-month-olds presented with a
series of syllables (2 mins)


Bidakupadotigolabubidaku
Transitional probabilities



Within-word: 1.0 (bida)
Between-word: 0.33 (kupa)
Then, they play repetitions of the
“words” and “nonwords”


If babies have learned the words, they
should prefer the nonwords.
And they do!
Language Development

In utero:


Can distinguish between vowels (/a/ vs. /o/)
Infants:



Can distinguish phonologically similar sounds at 2 months (they
lose this ability by 8-12 months)
Can pick up on cues to word segmentation
Can distinguish their native language from other languages
Language Development

How do we tell what
babies know?



Preferential looking
Sucking
Heart rate
Language Development


Babies prefer looking
at/hearing new things
When something is
interesting:


Heart rate slows
Sucking rate slows
Child Language Development

How do children get from being completely non-verbal to
being expert speakers?
1.
2.
3.
4.
5.
6.
7.
Can distinguish between vowel sounds (/a/ vs. /o/)- in utero
Can distinguish between all contrasts- from birth
Categorical perception of speech sounds (8-12 months)
Babbling: 6 months
One word stage: ~1 year
Two word stage: ~2 years (vocab is about 50 words)
Multiword utterances; gradually increase in complexity
Verb Learning

Two types of past tense
verbs:



Regular: talked, liked, hated
Irregular: ate, went, was
U-shaped curve of
language learning



Early: correct usage
Middle: overgeneralization
Late: correct usage
The Critical Period


If language learning doesn’t
occur before a certain time,
language will be impaired
Johnson & Newport (1989)



Age of Acquisition affects
ability to learn second
language
Genie
Pinker (NR)


Nicaraguan sign language
Deaf children