|Title:||Tracking and Coding of Referents in Persian and English: A conversational analysis||Add Dissertation|
|Author:||Hussein Shokouhi||Update Dissertation|
|Email:||click here to access email|
|Institution:||La Trobe University, Department of Linguistics|
|Linguistic Subfield(s):||Discourse Analysis;|
|Abstract:||This study shows that the system of referent tracking in conversation is two dimensional. A combination of interational and cognitive elements affects communication, conditioning the selection of certain grmmatical forms. Neither the cognitive factors alone nor the social ones per se have been found to determine the grammatical form of a given utterance.
The parameters discussed in the study are zero anaphora, nominals and pronominals studied in the light of the theory of Information Flow, developed mainly by Chafe (1980, 1987, 1993, 1994) and Du Bois & Thompson (1992)among others. These concrete items are measured against abstract notions such as givenness, newness, accessibility, topicality, identifiability, etc. To this end, over 10000 Intonation Units in Persian and English conversation in twelve different conversation settings were recorded, analyzed and compared with English.
The analysis of our data has revealed that although the cognitive parameters mentioned above are relevant to the presence or absence of different morphosyntactic forms (e.g. the given information being expressed in an attenuated i.e. pronominal or zero form), the appropraiteness of each form is also dependent on the interactional properties of the conversation such as 'turn-taking' mechanisms. Many occurrences of full NPs and pronominals have been found to be affected by conversational features. It is thus revealed that occurrences in full NP form are more numerous in the Persian data than in English. One significant reason is the use of formulaic politeness expressions. Zero anaphora has also been found to be pervasive in Persian conversation, but is constrained by factors such as turn-taking, repetition, and other interactional features.