Introduction
According to many linguists, it is known that speakers use full lexical noun phrases in their speech when a referent has not been previously mentioned in the discourse while a pronominal or zero nouns when a referent has just been mentioned. Du Bois refers to this pattern as Preferred Argument Structure, and he discusses the syntactic and pragmatic features associated with it. He defines it as “a preference in discourse for a particular syntactic configuration of linguistic elements, both grammatical and pragmatic”. As suggested by Du Bois, this notion of preferred argument structure can be better explained in terms of four constraints:
Grammatical form:
- Avoid more than lexical core argument.
- Avoid lexical As.
Pragmatics:
- Avoid more than one new core argument.
- Avoid new As.
These constraints illustrate the correlation between grammar and pragmatic. The claim is that, speakers tend to use lexical noun in intransitive subject position and transitive object position, but avoid lexical noun in transitive subject position. In addition, new information appears in intransitive subject or transitive object, but not in transitive subject role.
Du Bois claimed this tendency existed widely in discourse of all languages. The aim of the current study is to test this hypothesis based on a discourse analysis of oral and written Arabic narratives.
Arabic structure
Arabic language has a basic VSO word order. The standardized written Arabic is distinct and more conservative than all of the spoken varieties. As in other Semitic languages, Arabic has a complex morphology, that is, “root-and-pattern” morphology. In general, Arabic has three grammatical cases (nominative, genitive, and accusative). The cases are indicated by suffixed short vowels (/-u/ for nominative, /-a/ for accusative, /-i/ for genitive). These case markings are placed at the end of a noun or adjective. It is only used in written Arabic (Rydin, 2005).
The particular question of this study is:
What is the preferred argument structure in terms of number and type of arguments contained per clause in the oral and written Arabic narrative?
Methodology:
Subject
Data used in this study are taken from 10 Saudi Arabic speakers, 4 males and 6 females. Their ages range from18 to 35. A convenience sampling method was used to identify potential participants. Each participant completed the task in a quiet room in various locations.
Procedures:
Participants were asked to look at series of pictures that tell a story without words (see attachment). The subjects were given as much time as they needed to look at the pictures. They took between two to five minutes to identify the story. Then, their oral narratives were recorded with unlimited time constrain. After that, they were asked to write their story on a paper.
The data was then transcribed and coded for various discourse features. The unit of analysis considered was the clause. All clauses were divided into transitive and intransitive clauses. All the NPS were classified to A, O, or S. I also counted the number of arguments per clause and the different types of arguments (lexical, pronominal, and zero). Then all examples of each category were counted for all the data.
Result:
To answer the research question, I analyzed the oral and written Arabic narrative data according to the four constraints that suggested by Du Bois.
The pattern observed here is similar to that in other languages. There is relatively small number of more than one lexical argument per clause. This was shown only in the written narrative. Two lexical arguments per clause were shown in all the 10 subjects’ first clause in order to introduce the scene. For example, “an old man was riding his bike”, and “a man is driving his bike”. However, in oral narrative, all participants introduced the man only in the first clause by intransitive clause. For example, “there was a man. He was riding his bike”. However, the overall percentage of more than one lexical argument is relatively low.
The quantitative details of the second constraint data set are summarized in table 2.
As can be seen from the previous table, the oral narratives were more consistent with avoidance lexical As than written narrative. There was no any lexical A in the oral data. As will be shown in table 4, the majority of these written lexical As did not carry new information. Therefore, we can say that in written Arabic narrative, writers do not worry about being repetitive. As a result, they repeat the subjects lexically over and over. For example, one participants wrote “a man was riding his bike and the man saw an animal”¦..then the man carried the animal”¦the man went to the doctor”.
Another interesting observation can be made from table 2. In the table, there is a higher tendency in oral and written narrative to use zero nouns especially in the subject of transitive clause role. This pattern considers grammatical and preferable in Arabic. One factor increases the use of this pattern. In this data, there was only one participant in the story. Consequently, they did not need to repeat it to distinguish him from other participants in the story. An example of this pattern is “left his bike and took the rabbit then put him inside his jacket and left home”.
Again, the pattern in the oral narrative observed here is similar to that in other languages more than written narrative; there is avoidance of more than one new argument per clause. The use of more than one new argument per clause in the written data correlates with the number of lexical arguments per clause as shown in table 1. The reasons are the same also. It was mostly when the participants introduced the man and the bike in the first clause of their written narrative, but did not do that in the oral narrative.
The quantitative details of the fourth constraint data set are summarized in table 4.
After analyzing the data in table 4, there was no any new As in the oral narrative. There were few new As compared to the number of given As in the written narrative.
In summary, the four constrains of preferred argument structure are found in oral Arabic narrative data and written data to some extend since a small percentage of the A argument are lexical or new.