(toc) #title= (Table of Content)
The D-Structure
In recent times, the deep structure of the earlier models of transformational grammar has been changed to D-structure, and the surface structure to S-structure (representing the superficial syntactic structure of sentences). This is because the earlier models could not adequately handle certain characteristic constructions in natural languages. In order to provide a principled account of the syntax of these constructions, the D-structure level is posited. The D-structure and the S-structure are interrelated by a set of movement rules known technically as transformations. The D-structure does not solely determine the meaning of structures; meaning is realized at S-structure and is rediscoverable at the D-structure.
Movement of elements always takes place at the D-structure level. It is assumed, for example, that the subject position is always empty at the D-structure level, hence movement of an element from the object position to the empty subject position where case is assigned to it. Thus, we say that -role is assigned at the D-structure. This assignment can be illustrated by the following examples:
1. (a) You saw who?
(b) Who(m) did you see ti
Notice that in 1(b), who(m) is the D-structure direct object of see. In this position, it is assigned -role and Accusative case, but when it is moved to the empty subject position (ti represents the trace), it is assigned the Nominative Case. In example (2) below:
2. Johni was arrested ti
John is moved from its original position to the empty subject position where it is assigned the Nominative Case. It follows, then, that -role is assigned at the D-structure while the case is assigned at the S-structure.
We shall go on to show that the D-structure and the S-structure ated by looking at some transformational rules. Let us consider the rule of V movement.
Finite clauses contain a modal auxiliary such as will/would, can/could, shall/should, may/might, etc. Modals are generated under a finite I, and a finite I carries Tense (past/present) and agreement (person, number) features. Because these features are an intrinsic property of I, ause modals are generated under I, the relevant tense and agreement features are realized on the Modal. Thus, in (3) below:
3. John [I] [VP annoy me]
I will contain Tense and agreement features which are normally realized as a suffix on the Modal in I. However, when there is no Modal in I, these features are realized on the leftmost V of VP which is then inflected for the relevant tense/agreement features as in example (4):
4. John annoys me
where annoys carries the –s inflection which characterizes a third person singular present tense verb.
But how can tense/agreement inflections associated with I end up on a verb which originates as the head V of VP? The solution proposed by Chomsky is to suppose that there is a rule which assigns the elements of INFL to the initial verbal element of VP. Assume this rule to be Affix Movement rule. The rule involves affix movement in the sense te/agreement affixes which would normally be suffixed to the Modal in I end up suffixed to the leftmost V of VP. The rule of affix movement involves the attachment of an empty finite I to the right of V to form an inflected verb, and it will apply as represented in (5):
5. (a) John [I] [VP [V annoy] me]
affix movement
(b) John [VP [V annoys] me]
The result of affix movement is that the V, annoy acquires the relevant tense/agreement features normally associated with I, and the resultant inflected verb from annoys remains as the head V of VP.
However, a very different solution is suggested by other scholars. Here it is proposed that a finite clause where I does not contain a modal and hence is empty, the head V of VP moves into the empty I position by a rule of V movement (more specifically, V-to-I movement), and thereby
comes to acquire the tense/agreement properties associated with I, so becoming an inflected verb form. The representation will now be as in (6):
6. (a) John [I] [VP [V annoy] me]
V movement
(b) John [I annoys] [VP me]
The essential difference between the two analyses (5) and (6) is that under the affix movement analysis (5), the resultant inflected verb annoys remains within VP, whereas under the V movement analysis (6), the inflected V annoys ends up as a constituent of I. It is difficult to choose between the two rules because in English I and V are adjacent constituents, so that in either case the rule applies vacuously (i.e. the rule applies in such a way as to change constituent structure without producing observable change of word order). Therefore, irrespective of whether we adopt the affix movement analysis or the V movement analysis, we end up with the same overall NP V NP word order. For this reason, it is suggested that in choosing between the two rules, we must look at languages where I and V are not adjacent constituents, and where the two rules will not apply vacuously.
Although space will not allow us to discuss some more transformational rules, from the above discussion, we see that incorporation of movement rules into a grammar presupposes that there are two different levels of syntactic structure in grammars: (i) the level of D-structure which serves as input to the movement rules, and (ii) the level of S-structure which serves as the output of the movement rules. Clearly, S-structures are generated from D-structures, by the application of movement rules. But how are D-structures generated? We assume that the Categorial Component (or Base Component) of our grammar directly generates D- structures (or Base Structures) and that these then serve as input to a set of movement rules which convert or map them into the corresponding S- structures. The output of a movement rule is known as a derived structure.
CONCLUSION
In this unit, we have tried to show that it is important to posit a D-structure in syntactic analysis. We have discussed the functions and importance of the D-structure and how it interacts with the S-structure. You should familiarize yourself with all we have discussed to enable you to follow what will be discussed in subsequent modules.
Leave a comment behind