Stephen van Vlack

Sookmyung Women`s University

Graduate School of English

Introduction to Linguistics

Fall 2003

Answers for Cook and Newson, `Chapter 2`.

1. What are some of the major phases that Chomsky`s theory of language have undergone?

1. Chomsky`s theory of language started in 1957 with the publishing of book Syntactic Structures. In this book Chomsky outlines the basic idea of structural dependency. He also describes for the first time the underlying ideas of generative grammar. Phrase structure rules were introduced here as being separate from transformations, or movement within language. The next major stage in Chomskian theory came in 1965 with the publishing of Aspects of the Theory of Syntax. It was here that Chomsky further developed transformation theory by coming up with the concepts of deep and surface structure. This allowed for a better understanding of how language production works. Helping more in this direction was the outlining of the all important competence/performance distinction in language, which we know is apparent in the differences in the work of I and E-level linguists and has brought Chomsky a lot of criticism. After this Aspects/Standard theory time came the reign of Extended Standard theory which worked on limited and reanalyzing some of the transformational rules that were advocated in the Aspects approach. These changes eventually lead to the development of Government and Binding (GB) theory. The most important thing to come out of the GB movement was the introduction of the ideas of principles and parameters. The best examples of such principles are government and the concepts behind binding theory. GB theory was further constrained in the book Barriers, written in 1986. Barriers was primarily concerned with constraining the working of government while expanding its role in language and in doing so forms a bridge to the latest step in Chomskian theory, the Minimalist Approach.

2. What is the Chomskian model of language in principles and parameters theory?

2. The Chomskian model is best described in the T-model. In this model (for languages like English anyway), we start with the D-structure. Movement converts the D-structure to the S-structure. S-structure as we shall see more later is a real intermediary between all the levels. It must show all the movement from the D-structure and also dictates what is to be done in the next level. Changes for the D-structure to the S-structure are preserved in the S-structure via traces. These traces are necessary to let the PF and LF know what is intended with the utterance. From the S-structure the utterance goes into the areas of Logical Form (LF) and Phonological Form (PF). The LF-PF level works as a kind of cross checking filter to check that the form matches the meaning. It is hard to determine which one the utterance will go through first. Both are needed and the phonology needs to be able to go back and cross check what has been done with the semantics, so it seems that they need to be recursive in they can check each other. Once the utterance has been created and the proper checks have been made in the levels of PF and LF, then it emerges in its surface representation.

For Chomsky, syntactic structure plays the largest role in mediating the connection between sound (PF) and meaning (LF). LF and PF occur on the same level and function as intermediaries between language and the rest (the other cognitive systems) of the brain. It must also be understood that LF and PF are also syntactic in nature and that they work with the meaning and the sound in its syntactic context, but each are different systems of their own.

3. How does UG deal with languages like Korean which do not have movement rules in the syntactic component?

3. We have seen that English is highly reliant on movement to create different types of meanings. In fact, it is from studies of movement that modern syntax really began to develop. In the past, when confronted with so-called flat languages like Korean and Japanese linguists had a hard time incorporating them into a model where movement was central, since all the other levels were not at all integrated. The beauty of new UG models, like the T-model, is that this model also works with languages like Korean which do not have movement rules in the syntactic component by giving a larger role to the level of LF-PF. Since Korean does not have syntactic movement, there is no change in the syntax from the D-structure to the S-structure. This means that the very presence of the S and D-levels is subject to parameter variation in languages This does not mean that change and alteration, however, does not occur. For languages like Korean, syntactic information is encoded morpho-phonemically (in the PF and LF) and Chomsky believes that the vast majority of this encoding must take place in the LF. Korean does have movement, but this movement is in the LF and is not syntactic movement as we see in English. This is possible now because the model shows a lot of integration among the parts even though they each have their own special jobs to do and configuration.

4. What is theta-theory and how does it work?

4. Theta-theory is the idea that all nouns in clauses are given semantic roles, such as GOAL, EXPERIENCER, AGENT, etc.. These roles are given to the nouns by the verbs in the clause. The theta-roles are given out in much the same way that syntactic information is. Chomsky believes that information about the roles that verbs put on the nouns within the same clause is encoded in the mental lexicon along with the entry for the verb. When the verb in question is fired in the brain theta-role assignment as well as syntactic information is accessed and used to create D-structure utterances. The appropriateness of the theta-roles are later checked in the LF component.

5. What is INFL and why do we need it in the theory?

5. INFL is short for inflection. It is in the theory as a single category despite the fact that there are many different forms for the one category. These forms are auxiliaries, infinitive to, and TENSE and AGREEMENT features. These are combined into one category because all the different forms function the same syntactically. The positing of INFL is important for the overall theory of syntax because it appears to be more correct. We know it is more correct because the category of INFL plays an important role in many different syntactic operations. This is best seen when we examine the role of INFL as a governor. In modern syntactic theory INFL is one of the highest branching nodes of the tree. This means that INFL governs the noun in the subject position of the clause. We need now to make a distinction between finite and non-finite clauses. Clauses that include TENSE and AGREEMENT are classified as finite. Clauses that do not have these features (infinitival clauses) are non-finite. Look at the examples below.

A) [The pigs ate their breakfast of ham and sausage greedily]. (This is finite.)

B) [The armadillo wants [ to kiss me]]. (This, the second clause, is non-finite.)

C) [The dog wants [him to give him a bone]]. (This, the second clause, is non-finite.)

In sentence B, there is no subject. This has lead syntacticians to come to several conclusions. The first of these is that only finite INFL can govern the subject and therefore determine its case. We know this because the subject should be given the case of nominative by the governor, but utterances like C) show that this is not the case with non-finite INFL governors. Do not worry if you don`t get all of this now. We will be coming back to this in much greater depth later.

6. How does pro-drop function as a parameter of language (think about INFL)?

6. Pro-drop functions as a parameter of language in that all languages are either pro-drop (Korean, Chinese, Spanish) or non-pro drop (English, French, German). Pro drop languages, as the name implies, are able to have subjectless clauses throughout. Non-pro drop languages, on the other hand, are not normally able to drop their subjects. Generative grammar says that pro drop languages have an empty category in the subject position. Try to get used to the idea of empty categories because we will find a lot more as we get deeper into this. Syntactically speaking the subject position is not really empty at the S-structure level. It is filled with something called pro (little pro). Now, that is simple enough, but let`s take it one step further.

We looked before at INFL. INFL as it turns out, is not a proper governor (as we saw above) according the Principle of Proper Government. (Remember we said above that non-finite INFL cannot assign case to subjects). Only lexical categoreis can be propoer governors, right?

Now, when we combine the idea of Proper Government with the Empty Category Principle we find out a lot. The Empty Category Principle says that empty categories must be properly governed. In pro drop languages we have empty categories in subject position (i.e., a position governed by INFL) this means that finite INFL must be a proper governor in pro drop languages. If this were not so, then such utterances as Spanish

Voy a mi casa. Go(1st pers. sing. form) to my house (I am going to my house.)

would not be possible. Finite INFL is not a proper governor in non-pro drop languages because such languages cannot have empty categories in a position governed by INFL. What we are doing here is using the idea of proper government to account for pro drop in languages. In this way the value of finite INFL as a proper or improper governor is another parameter. Taking this one step further we can see that different values for the concept of proper governors either allows pro drop or keeps it from happening.

7. How does Binding Theory work in language? Why is this important?

7. Binding Theory is really quite easy. According to the theory there are three different types of categories that work in pronoun situations. The first of these is the referent or referring expression, then we have the pronominal, and finally there is the anaphor.

A) The armadillo scratched him.

B) The armadillo scratched himself.

In both sentences the armadillo is the referent. In sentence A, him is a pronominal. In sentence B, himself is an anaphor.

The simplest thing to do is to place these is a hierarchy based on freedom. A referent is always free. A pronominal is free in its local domain. An anaphor is bound in its local domain. It is quite simple and I don`t think I need to give a whole bunch of examples here to support this since we did that in class. Binding Theory is important because it was the first attempt at using relative distance to determine structural relationships as well as semantic ones. It went very far in paving the way for the road ahead to Minimalism.

8. How do the core and periphery of language relate to each other?

8. The core of language is basically the i-language that Chomsky and his followers work. At the same time, however, Chomsky says that language is made up of more than just the core. There is also the periphery. The periphery is the part of that language that is not necessarily part of the core because it is not systematic or productive. One of the biggest example of peripheral language in English would be irregular verbs and nouns. These verbs and nouns are the result of historical mishaps. Change in language (and languages are always changing) does not occur regularly. Chomsky feels that he does not need to account for such peripheral parts of language as they run contrary to the systems which make the language productive. He sticks to the core and everything that does not fit the core is regarded as peripheral.