Universal Grammar (UG)

Universal Grammar (UG): an innate, genetic endowment of language-specific knowledge consisting of the principles and parameters of language. An idea promoted by Chomsky (more developed notion of LAD)

Principles of Language: abstract principles that permit or prohibit certain structures from occurring in all human languages

Parameters of Language: systematic ways in which human languages vary, usually expressed as a choice between two options

Let us recall the basic character of the problem we face. The theory of UG must meet two obvious conditions. On the one hand, it must be compatible with the diversity of existing (indeed, possible) grammars. At the same time, UG must be sufficiently constrained and restrictive … to account for the fact that each of these grammars develops in the mind on the basis of quite limited evidence … What we expect to find. Then, is a highly structured theory of UG based on a number of fundamental principles that sharply restrict the class of attainable grammars and narrowly constrain their form, but with parameters that have to be fixed by experience…the languages that are determined by fixing their [parameter] values on way or another will appear to be quite diverse…yet, at the same time, limited evidence, just sufficient to fix the parameters of UG, will determine a grammar that may be very intricate and will in general lack grounding in experience in the sense of an inductive basis. (Chomsky 1988, Lectures on Government and Binding, 3-4)

The logical problem of language acquisition

or

The “poverty of the stimulus”

· Without a genetic endowment, first language acquisition would be impossible because the input data are insufficiently ‘rich’ to allow acquisition to occur with purely inductive means.

· In most cases, the (first) language-learning environment does not provide evidence concerning the well-formedness of an utterance

· Even when information is supplied that an utterance is ungrammatical or inappropriate, the information on how to modify the utterance is not usually supplied.

· Explicit teaching and correction has little to no effect on child language acquisition

· The child largely hears only positive linguistic evidence (no information about what is ungrammatical)

o How do they learn that what they haven’t heard is coincidence or b/c it is impossible

Problems with only hearing ‘positive’ evidence as input (White 1985)

Given the following input:

Peter stole something

What did Peter steal __?

The policeman believes that Peter stole something.

What does the policeman believe that Peter stole __?

The lawyer says that the policeman believes that Peter stole a watch.

What does the lawyer say that the policeman believes that Peter stole __?

How does the Learner know the following are incorrect?

Mary likes the children in her morning class.

*Which class does Mary like the children in __?

His love of weapons frightened her

*What did his love of __ frighten her?

The workers believed the story that the owner would close their factory.

*What did the workers believe the story that the owner would close__?

They wondered whether he would declare bankruptcy.

*What did they wonder whether he would declare __?

Answer: The Principle of Subjacency

Subjacency Principle: A constituent such as a WH-word may only be moved across one bounding category

Bounding Category Parameter: in English Sentences (S) and Noun Phrases (NP) are bounding categories

Only crosses one boundary:

What_i [did Peter steal t_i ]_S?

Only crosses one boundary at a time, since it moves to COMP

What_i [does the policeman believe [t_i that [Peter stole t_i ]_S]_CP]_S

What_i [does the lawyer say [t_i that [the policeman believes [t_i that [Peter stole t_i]_S]_CP]_S]_CP]_S

Crosses more than one boundary:

*Which_i class [does Mary like [the children in t_i]_NP] _S

*What_i [did his [love of t_i]_NP frighten her] _S?

*What_i [did the workers believe [the story that the owner would close t_i]_NP] _S

*What _i [did they wonder whether [he would declare t_i]_S] _S

Another Example of “Poverty of the Stimulus”

Haegeman (1994) Introduction to Government and Binding Theory

Input the child receives:

I think that Miss Marple will leave.

I think Miss Marple will leave.

This is the book that I bought in London.

This is the book I bought in London.

Who do you think that Miss Marple will question first?

Who do you think Miss Marple will question first?

The child might conclude that the conjunction that is optional.

*Who do you think that will be questioned first?

Who do you think will be questioned first?

· How can the child infer this from the input?

· Gap between the data we are exposed to and our knowledge.

Answer:

There is a general principle that does not allow a subject to be moved from a position to the immediate right of a conjunction.

The PRO-drop Parameter

Italian	English
Omitted Subjects:
Lei ha invitato Louisa a casa.	She invited Louisa to the house.
Ha invitato Louisa a casa.	Invited Louisa to the house.
Subject in post-verbal position:
È arrivato Gianni	*Is arrived John
Ha telefonato sua moglie.	*Has telephoned your wife
Che Louise non partirà è chiaro.	That Louise will not leave is clear.
È chiaro che Louise non partirà.	*Is clear that Louise will not leave.
*Ciò è chiaro che Louise non partirà.	It is clear that Louise will not leave.
Dummy Subjects:
*Ciò piove.	It is raining
Piove.	*Is raining.
Movement of Subject of Subordinate clause:
Chi credi che abbia telephonato?	*Who do you think that has telephoned?

This last example seems like a violation of the principle constraining subject movement:

But consider the two possible derivations:

Chi credi che ___ abbia telephonato?

Chi credi che abbia telephonato ___ ?

Now Consider the data from Spanish and French.

Spanish	French
Omitted Subjects:
Baila bien	Elle dance bein
“(she) dances well”	*Dance bien.
Subject in post-verbal position:
Llego Maria ayer a los doce	Marie arrivait hier a midi.
“Mary arrived yesterday at 12”	*Arrivait Marie hier a midi.
Dummy Subjects:
Me parece que Juan tiene hambre.	Il me semble que Jean a faim.
“It seems to me that John is hungry”	* Me semble que Jean a faim.
Movement of Subject of Subordinate clause:
Quien dijiste que vino?	Qui dis-tu viendra?
“Who did you say came?”	*Qui dis-tu que viendra?

Parameters and Markedness

· Unmarked parameter settings (default settings) are always the most conservative or restrictive

· Parameters converted to marked setting by experiencing utterances (positive input) consistent with the marked setting

Parameter for preposition + object movement:

+ Pied-Piping – Preposition Stranding (unmarked) Spanish

+ Pied-Piping + Preposition Stranding (marked) English

Pied-Piping:

From where did you get that fish?

With which friend does Dorothy live?

Preposition Stranding:

Where did you get that fish from?

Which friend does Dorothy live with?

Parameter for co-reference of reflexives:

Local Binding (unmarked) English

Long-Distance binding (marked) Japanese

[The professor_i saw himself_i.] _S

[The student_i knew that [the professor_j saw himself_j.] _S] _S

*[The student_i knew that [the professor_j saw himself_i.] _S] _S

UG and SLA

If we assume that second language grammars are natural grammars, then SLA data inform our thinking on linguistic principles and parameters.

· If we find that SLA follows UG constraints, it would tend to support the idea of principles and parameters

· If we find that SLA does not follow UG constraints, we can:

o Assume that SLA does not have access to UG (but that first language acquisition still does).

This position has been called The Fundamental Difference Hypothesis

o Attribute the results to methodological problems.

In that case we should re-do the study and then replicate those findings.

o Assume that our understanding of a certain parameter of principle of UG is false.

Note that the descriptions/definitions of principles and parameters change constantly

o What else?