Stephen van Vlack
Sookmyung Women`s University
Graduate School of English
Introduction to Linguistics
Fall 2003
Answers for Cook and Newson, `Chapter 5b`, pp. 173-188.
1. What are the major grammatical functions (GFs) and where are they located?
1. The major grammatical functions (GFs) are subject and object. According to the regulations set by x-bar syntax, which serves to regularize syntactic/structural phenomena, these GFs only occupy certain positions. It is their occupation of certain locations which actually serve to define them. Subjects are located at least at the level of D-structure in the specifier position of the VP (NP X). This means that part of being a subject is being in the Spec VP or IP position. Objects are, of course, found in the complement position of categories, especially the VP (NP X`). These GF positions are the positions that these grammatical functions are inserted into in the D-structure. We know that they often move for the S-structure. It is still an interesting claim that certain GF occupy or are inserted into a specific position on a universal level.
2. What is the difference between A-positions and non-A-positions?
2. The difference between A-positions and non-A-positions basically comes down to which one can take arguments and which one cannot. A-positions are argument positions; i.e., positions in the tree in which arguments are found. They are Spec VP, Spec IP, Comp VP and Comp PP. A-positions correspond, not surprisingly, to GF positions. Non-A-positions are obviously positions in which no arguments are allowed to be projected. The most common non-A-positions are Comp IP and Spec CP. We know that Comp IP is occupied by the VP, which is a predicate and not an argument and Spec CP is a position left empty unless something (like a Wh-word) is moved into it.
3. How do A and non-A-positions relate to theta-marking?
3. In addition to A and non-A-positions we also have what is called theta-()-positions. Theta positions are positions in which theta roles are assigned. Theta-roles can only be assigned to sisters. When theta-roles are assigned to sisters of a head is called internal theta-marking. When theta-roles are assigned to sisters of X` is called external theta-marking. There is a strong, but not reciprocal relationship between A-positions and theta-positions. All theta-positions must also be A-positions. All A-positions, however, do not have to be theta-positions. Spec IP, for example, is an A-position but is not a theta-position, the idea being that the subject originates in Spec VP position (a theta-position) and moves to Spec IP. A-positions that do not receive theta-roles are always ones that are related to the GF subject. This is an important point to remember when we get to movement.
4. What does the Extended Projection Principle claim?
4. The Extended Projection Principle claims that all clauses must have subjects. This means that unlike objects, which may or may not be required based on the c-selection of the verb, subjects are always required. This accounts for the use of pleonastic pronouns as subjects in such sentences as...
A. It was a terrible day for the armadillo.
B. There were three babies in the hallway.
C. It seemed that the elephant was crabby.
Chomsky uses the presence of such pronouns to argue that subjects are necessary in English. The basic idea is that it and there are inserted into the subject position even if a semantic subject is unnecessary in order to fill the position. These pleonastic subjects are not theta -marked because they are inserted straight into the Spec IP position (a non-theat-marking position) and not the usual subject position, Spec VP, which is a theta-marking position.
5. What are the main differences between functional and lexical phrases?
5.The main differences between functional and lexical phrases are many. First of all lexical phrases (with the noticeable exception of prepositions) are what we call open categories. There is no fixed number of lexical units which can belong to any given lexical phrase. The number or types of nouns and verbs is potentially unlimited. Functional phrases, on the other hand, are limited in number and may not even terminate in an overt word-level representation. Lexical phrases (what is known as content words in the applied linguistics literature) are also stressed while functional phrases are unstressed. Prepositions are the noticeable exception here. In fact, prepositions seem to be a special kind of lexical phrase and might be better placed a straddling the fence between lexical and functional.
There are many other little differences, but the last difference I will talk about here is the fact that functional phrases are linked to parameters while lexical phrases are not. This is an indication of the more abstract nature of functional phrases. Functional phrases and their use is often dictated by parameter settings. It is here then that we find the greatest amount of cross-linguistic difference.
It may be helpful to return to an idea we mentioned a few weeks ago in which it was mentioned that all languages might be the same in their lexical phrases. That is, not only the lexical phrases themselves, but also their c and S selection might be universal. The actual forms of the lexical words will obviously be different, but they might all actually be the same in every language. Following this idea, it is the only in the functional phrases (the domain of parameters) that we find actual differences in languages. I think this is a bit far-fetched, but is intriguing in any case.
6. Why is the positing of a D-phrase better than the N-phrases we saw earlier?
6. There are several reasons why the positing of a D-phrase is better than the N-phrases we saw earlier. The most obvious of these is that it will make the entire system follow the rules of x-bar syntax. A Determiner Phrase can be filled with a wide variety of different lexical items, from articles to pronouns and nouns as indicated in the examples below.
A. the man
B. his brother
C. the armadillo`s book
The readings we get for B and C is that the underlined elements are the heads. If this is so, then the article in A must also be the head. This brings regularity back into the system because with the DP analysis we are putting the heads first not last. We know English is a head-first language, so this must be right. Under the old NP analysis the heads came last, which seems to be wrong for English. Additionally, the DP analysis allows us to comfortably deal with such phrases as in B and C very elegantly and without having to posit lots of extra rules. Although the DP analysis has not been accepted by all working linguists yet, it seems to have many positive points.