Week 11 Flashcards
Children’s mental grammar have
____ before they start producing multiword utterances
syntax
Do kids have adult-like grammar?
If the universal grammar (nativists) people are right then what kind of grammar would we observe in young kids?
- kids have pretty sophisticated grammars by age 5
- if the universal grammar (nativists) people are right, we should be able to observe adult-like grammar in young kids
What is the Continuity Hypothesis?
(Pinker, 1994)
- it suggested that in the 80’s, grammatical development is continuous (that at birth, the child’s mind includes all the structures that make up adult grammar)
- the stuff (structures, syntax etc…) of the grammar is there at birth
- born with a lot of syntactic knowledge already
Explain the longitudinal study about Eight “relational meanings” . (Brown, 1973) ex: agent + action Daddy sit agent + object Mommy sock action + object
Is this consistent with the continuity hypothesis?
- Adam, Eve and Sarah
- sent lab assistants to their houses once a week for two or three years
- naturalistic data of them interacting with their families in their homes
- in the early utterances when they first started combining words, they observed all these combinations in the table
- is this consistent with the continuity hypothesis?
- brown is saying that when they first start combining words, they are not using syntax
- they are actually trying to use semantic principles (meaning) to combine words
- NOT consistent with the continuity hypothesis!
What is the truncated clause theory?
(Radford, 1990; Guasti, 2002)
Is it consistent with the continuity hypothesis?
- a simple sentence
- theory says that they are obeying syntactic principles by adult-like syntactic principles but that fully detailed tree hasn’t yet developed
- their utterances are grammatical in respect to their mental grammar but the extra levels of structure don’t get turned on until the kids get older
-the Truncated Clause theory is consistent with the continuity hypothesis
Do kids have adult-like grammar?
• kids’ performance is constrained by processing limitations (Bloom, 1990; O’Grady, 1997)
Is this consistent with the continuity hypothesis?
- the reason they say things that are not grammatical, they have processing limitations
- domain-general constraint
- they have the competence but its a performance limitation
- memory ability is limited
- they get the content words out but not the functional words
-consistent with the continuity hypothesis
Discuss the study about Evidence from Production
.
(Cazden, 1968; Maratsos, 1982)
• M: 74 hours from 21 children (2;9 - 4;0)
• C: 150 hours from Adam, Eve, Sarah
• looked for syntactic category errors
- a verb in a place where a noun goes —> that would indicate that their mental grammar is not adult-like
- in that 224 hours, they only made 10 errors
ex: (noun used as verb) - You have to scale it first
- Mummy trousers me
-not using words in the wrong categories
Discuss the Evidence from Production study.
(Bloom, 1990)
• 14 children, 1;6 - 2;10
• looked at adjectives & nouns
-Kids used adjectives in two different ways
How can we interpret this pattern?
• kids used predicative adjectives with common nouns, pronouns, proper names
ex: pillow dirty, girl bad, I sleepy
• kids rarely (
What do the production results tell us? If this was just about semantics then…
-we wouldn’t expect to see these results in the difference between the adjectives, proper names and pronouns
What is a principle?
the things that are true of every language
What is a parameter?
-the systematic ways that languages vary
The x-bar theory. principle or parameter?
principle
Every sentence has a subject.
principle or parameter?
principle
In some languages, the subject can be null (i.e. unpronounced).
principle or parameter?
parameter
How would a child learn that Italian is null?
a) assume null subjects are allowed; observe null subjects in the environment; assumption confirmed
b) assume null subjects are forbidden; observe null subjects in environment; revise assumption
B
-the default setting is to not allow NULL subjects
Discuss the study about A competence account
.
(Hyams, 1987)
• English-acquiring kids first allow null subjects
(set parameter to [+null])
- kids start by allowing null subjects
- they learn from the positive evidence that comes from sentences with expletive subjects (its raining)
- ‘it’ is just a place holder because english has to have subjects
- its by observing this evidence even though there’s no theta role assigned to that position , we still have to speak subject in that position because english is [-null]
Evidence against the competence account.
• Expletive subjects appear in kids’ productions several months after null subjects disappear (Ingham, 1992)
• Comparing English and Italian kids’ productions
(Valian, 1991)
-if english kids started with a default parameter setting the same as italian kids then their production would be the same
-english kids are dropping the subject sometimes but not as much as italian kids
• English 70% overt subjects
• Italian 30% overt subjects
Discuss the study about A performance account.
Bloom, 1990
- English-acquiring kids don’t allow null subjects (i.e. set parameter to [-null])
- processing limitations impede kids’ production of subjects
Evidence for the performance account
(Bloom, 1990)
Restrict study to sentences where the verb is in the past tense.
What does the performance hypothesis predict?
- CHILDES study of Brown’s data
- Adam, Eve, Sarah, 1;6 to 2;7
- sentences with past tense verbs
- measured VP length in words
-kids will drop subjects in longer sentences
How do kids form yes-no questions? (Crain & Nakayama, 1987)
2 ways but which one do we actually use?
1) use a linear order strategy (start at the beginning and go along until we see our auxiliary verb and move it to the front)
2) structure dependent operation (look for the thing thats in C and move it up to T)
-no languages use a linear operation
Discuss the study about How kids form yes-no questions? (Crain & Nakayama, 1987) What were the results?
- when you’re trying to elicit certain structures from kids, it would take a while
- can’t ask kids to say if a sentence is grammatical or not
-technique used: picture or toys
- a boy who is watching mickey mouse,
“ask jabba if the boy who is wathcing mickey mouse is happy?” (talk to a toy in case they are afraid of the adult)
-by 4-5 year olds, they are accurate in forming sentences
-but 3.5 year olds are not accurate in forming sentences
What kinds of errors do kids
make?
(Crain & Nakayama, 1987)
-prefix errors and restart errors
What kind of operations are kids using to form yes/no questions?
structure-dependent operations
Binding theory: what is the restriction on pronouns?
-the pronoun must not be bound in its binding domain
- Emily likes her.
- She likes Emily.
Why are these coreferences ungrammatical?
- she/her can’t be emily (can’t corefer)
- universal rule (every language)
What is principle C of the binding theory?
R-expressions must be free
Do kids know Principle C of the binding theory?
(Crain & McKee, 1985)
• truth-value judgment task (TVTJ)
• toys act out scenario, puppet observer describes scenario, child evaluates puppet’s description
• “While he was dancing, Kermit ate pizza.”
• “He ate pizza while Kermit was dancing.”
- the truth of that sentence lets us know how they interpreted it
- if kermit’s right, he gets a cookie
- if kermit’s wrong, he gets broccoli
- they interpreted that ‘he’ refers to ‘kermit’ in first sentence
- but ‘he’ in second sentence refers to George not kermit
- kids said that sentence two was false
- if they say true to the first statement then that tells us ‘he’ corefers to ‘kermit’
- 3 year olds do know the binding theory
- only way to learn that this is ungrammatical is from negative evidence
- very hard to learn from the environment
Can kids learn from negative evidence?
yes but it takes time
middle ground between emergentists and nativists
