Transformational grammar was the starting point of a rather influentional theory of sentence meaning known as the Deep Case Theory or the Frame Theory. It should be immediately noted that this theory finally fused with the Valence theory.
The main idea of the Deep Case Grammar is that each verb has a specific set of functional positions for nouns. This set is called a frame and the functional positions are termed deep cases. The term “deep case” was used to indicate that only the meaning of certain syntactical relations is represented in the frame. Morphological cases are considered to be superficial, accidental and only indirectly connected with the deep cases as possible forms of the latter.
We can not but recognise that this description of sentence meaning is very similar to the valence theory and especially to the logical valence. The main difference between the Deep Case Theory and the Valence Theory is in the origins and the philosophical background. The Valence Theory was a further development of the Model of Members of the Sentence, while the Deep Case Theory was a result of searching for kernel structures of the Transformational Theory. Trying to discover kernel structures Ch. Fillmore found that they can not be made up by immediate constituents as the latter are defined in the IC Theory. Instead his material made him conclude that the kernel structures first should be meaningful and second consist of a predicate and a number of nominal elements whose quantity and syntactical meaning depend upon the lexical meaning of the predicate. At the same time he found that the syntactical meanings of the nominal elements could be named by the terms used for Latin cases. His discoveries permitted him to develop a semantic theory of a sentence. According to his theory the meaning of a sentence can be treated as a frame consisting of deep cases, or nominal elements having specific meaningful functions, and the predicate which determines this frame. The deep case frame, according to Ch. Fillmore, is a description of some state of affairs. The formal structure of the sentence is a result Of formal or in his terminology – surface – operations aimed at making the frame percievable for communicators. Thus the semantic structure of the sentence represents the structure of what we want to speak about and is only indirectly reflected in the form of the sentence.
We can see that this short description of the Deep Case Theory shows that it derives the meanings of the whole sentence and its components directly from our knowledge of the world. Because the general meaning of a sentence is determined by the type of the deep case frame and the latter is defined by the meanung of the verb, the verb meaning or more exactly the structure of the situation named by the verb is the general meaning of the sentence and the roles of the components of this situation are the syntactical meanings of the nominal elements of the sentence.
This makes the number and meanings of the Deep Cases unlimited. If we compare the roles of the first position in the case frame of the verb to open in the following four sentences: “Nick opens the door”, “The wind opens the door”, “The key opens the door”, and “The door opens ”, we can see that their roles in the situations described in the srntebces are different. In the sentence “Nick opens the door” the first noun is the Doer who ddoes the action consciously and can use an Instrument to perform the action of opening; in the second sentence (“The wind opens the door”) the first noun names a phenomenon that cannoy do anything and cannot use any Instrument, thus it cannot be treated as a Doer; in the third sentence (“The key opens the door”) the first noun names the Instrument used by the Doer; and in the fourth sentence (“The door opens ”) we find that the first noun names the Patient of the action. As a result in these sentebces we find: Agent in the first sentence (“Nickopens the door”); Elementative in the second (“The windopens the door”): Instrument in the thied (“The keyopens the door”); and Patient in the fourth (“The dooropens ”).
The same is true for the second position, in which we expect Patients. If we compare the following sentences: "Nick broke a cup", "Mike opened the door", "Jack built a house", "Mary read a book", we can notice that the nouns in the second position have quite different roles in the situations described in the sentences.
The thing named by the word cup in the sentence "Nick broke a cup" undergoes a radical change as a result of which it disappears; the thing named by the word door in the sentence "Mike opened the door" does not cangne so radically, it only changes its position; the thing named by the word house in the sentence "Jack built a house" does not undergo any changes as a result of the action of building because it appaers after the action is completed; the thing named by the word book in the sentence "Mary read a book" does not change in amy way, rather it is Mary that changes as a result of reading.
If we follow the logics of the Deep Case theory we should say that in the second position of these sentences we found four different deep cases (and one more in the first position because the features of Mary in the sentence "Mary read a book" differ from those we found in the sentences “Nick opens the door”, “The wind opens the door”, “The key opens the door”, and “The door opens”.).
This all means that the Deep Case Theory based on the assumption that the general meaning of a sentence is determined by the type of the deep case frame and the latter is defined by the meanung of the verb, the verb meaning or more exactly the structure of the situation named by the verb is the general meaning of the sentence and the roles of the components of this situation are the syntactical meanings of the nominal elements of the sentence produces an infinite number of deep cases, more to this, the number of cases and their meanings largely depend upon the desision of the grammarian: one may consider that the differences noted in the paragraph before are very important. Someone else might think that they are not. And some other might believe that it is better to group cup and house together, and oppose them to book and door, also grouped together.
This absence of limitations allows to explain many specific occasion of thew formal and semantic properties os sentences but it does not permit develop a general explanatory system. Still the idea that the sentence meaning is a special structure independent of the formal structure of the sentence proved to be fruitfull and is employed nowadays practically in all semantic theories of syntax. One of them will be briefly described in the next section