Stephen van Vlack
Sookmyung Women`s University
Graduate School of English
Introduction to Linguistics
Fall 2003
Answers for Cook and Newson, `Chapter 6a`, pp. 189-215.
1. If Move a allows all movement, how, then, is movement constrained?
Movement is constrained by Universal properties, such as Structural Dependency. Structural Dependency constrains movement by ensuring that only constituents can be moved. Movement is also constrained rather tightly by parameters. These parameters explain how movement is different for different languages and may or may not even take place in certain languages. Korean and Japanese have long been thought to be flat languages. They are called this because they do not seem to need the level of X` in their descriptions. They do not involve movement in the same way as a language like English. Chomsky, however, posits that movement does occur in Korean and Japanese, but at the level of LF. We will try to come back to this later.
2. What triggers NP movement?
NP movement is triggered in English by the presence of certain morphological structures. NP movement in the passive is triggered by the presence of passive morphology and an optional empty subject as well. Often passive sentences have an empty subject in the D-surface level and the NP object moves up into that position before moving on the AGRP Spec position, but in some cases there is an overt subject.
Cil Cuillan was killed by the one-eyed giants.
The triggering in this sentence is the same as in utterances with an empty subject. It is the morphology [was + Ved + (by)].
3. How does theta-theory relate to movement?
Theta-theory plays an important restrictive role in NP movement. Let`s look at the sentence below.
Kim Young-sam was beaten senseless.
The NP (Kim Young-sam) originates in the VP Comp position in the D-level. There is an empty NP subject. The object NP is awarded a theta-role <EXPERIENCER> before it moves up into the empty weak subject position. We know that this must be true because in the S-level Kim Young-sam is still the EXPERIENCER. This means that NP movement must occur in relation to theta marking. The NP can only move from a theta-position (VP Comp) to a non-theta-position (AGRP Spec). If it first moved to the strong subject position (VP Spec), then it would get two theta-roles, which we know is a violation of the theta Criterion. So, what we find out is that NPs must move from a theta-marked to a non-theta-marked position.
4. How does wh-movement differ from NP-movement?
Wh-movement is different from NP-movement in that wh-movement is what is called Non-A-movement while NP-movement is A-movement. NP-movement involves moving NPs from A-positions to other A-positions (Non--marked ones). In Wh-movement, the Wh-element is moved from an A-position to a non-A-position (CP Spec). We know that Wh-elements move into CP Spec because we can make a sentence where everything else is filled. Look at the sentence below.
What if I were to kill the dancing armadillo?
The C-element if is certainly in C position, the only position that the Wh-element an occupy is the CP Spec slot. This also follows the idea that categories occupy certain positions. We have already seen that Spec is one of the positions that NPs often occupy, so we are OK here.
5. What triggers Wh-movement?
Wh-movement is triggered by a parameter which applies to the CP. It has been proposed that the functional phrase CP (remember functional phrases are the domain of parameters) has a value that is either + or - interrogative. Interrogative is basically a fancy word for question. So, basically languages in which CP is marked as +INTER will have question words or particles move into CP. Languages like Korean that are -INTER do not need question words or particles to move into CP. In the book they call this parameter +/-Wh and because of that we will change our formalism to Wh instead of INTER with the note that I think +/-Wh is too language specific. So the basic idea is that CP is marked as being either + or - Wh. Again a +Wh complementizer requires a Wh word. Now, in order for this to work all complementizers need to have this value +/-Wh encoded in their lexical entries. As a Wh-word is projected from the lexicon into the structure for an English speaker it then has to move forward. Now we have a slight problem here because Wh-words move not into C, but into Spec CP position. If the Wh-word in is Spec CP position how can it satisfy the Wh requirement of the CP? In order to make this work an arrangement called Specifier-Head Agreement has been proposed. Specifier-Head Agreement basically says what its name implies. It proposes that there is agreement between the head of a phrase and the element in specifier position. If we look at the behavior of subjects (occupiers of Spec AGRP) in relation to agreement features, we see that this proposal has validity. It also ensures that moving Wh into Spec CP satisfies the Wh requirement of CPs in English.
6. Why do subjects have to move?
Subjects have to move in English in order to avoid the no crossing branches constraint on trees. We know that subjects originate in the Spec VP position. If the subject were to stay there then we have a problem with helping verbs.
a. John killed the armadillo
b. Jill did smack her lazy boyfriend.
In sentence a, it is hard to tell if the subject has moved or not. The features of Agreement and Tense move down to be marked on the verb. Sentence b, however, offers clear proof that the subject must move. The AGR position is filled by the helping verb did, yet the subject still precedes it in the sentence. Since we cannot cross branches in our trees the only way for this to occur is if the subject has, indeed moved from Spec VP to Spec AGRP position. Because we are looking to find regularity in all the proposals we make, then it makes sense (is necessary in fact) to propose that the subject always moves from Spec VP to Spec AGRP.
Additionally, subject movement is based on parameters. This judgement was arrived at based on the fact that subject movement does not occur in all languages. Linguists, therefore, conclude that the movement of the subject in languages like English is triggered by a parameter which forces the subject to move.
7. In English, if not the verb, what moves?
In English it is not the main verb that moves. What moves is a helping verb or auxiliary. In effect, verbs are blocked form moving in English. This is obvious from the observation that whenever a verb might be called on to move an auxiliary is inserted to do the job. Look at the possible conversation held by three friends who all know John
B: Did John really shoot himself in the foot?
In A, the verb is marked with the Tense and Agreement features, but in B where a question is being made, a dummy auxiliary (did) is used. Once more, since did and not shoot caries the Tense and Agreement features we assume that do (inserted into TP bounces up and picks up Tense and Agreement as it does so. The verb does not move because it does not have any tense or Agreement marked on it.
8. What parameter setting accounts for verb movement in English?
The parameter setting that accounts for the fact that verbs cannot move in English is strong or weak AGRP. A strong AGRP allows a verb to pass through it, as in French, but a weak AGRP does not allow a verb to pass through it. Since the value of AGRP in English is weak, then movement is blocked. A dummy verb (an auxiliary is created to do the job that the verb can`t do).