Abstract:

This dissertation focuses on the use of the article the in English, in order to answer two questions that have long puzzled linguists. First, why do speakers use the article the in a noun phrase when they do? Second, what elements in discourse enable the hearers to interpret the referent denoted with the as it is? The dissertation argues that English speakers use the article the to introduce a concept when it shall be conceived as a "dependent concept" specific to the discourse world. In the meantime, the hearers interpret the referent based on "genre-specific" conventions triggered by the article. It proposes that speech participants, in order to interpret the entity denoted by the word the in each communication, would need a communicative assumption to state how the content of the communication (i.e., the anticipated discourse world) may relate to themselves and to the reality.

In addition to the above proposal, the dissertation presents two studies. The first one reports data collected from 1417 definite noun phrases in a corpus of various genres. It was found that not all referents denoted with the in English are "anaphoric" or "familiar". Approximately half of the definite noun phrases with the found in the corpus do not have an explicit previous mention in the discourse. The study describes various types of these noun phrases, their frequencies, and some of their characteristics.

The second study reports data conducted from a web search. It presents frequency data to show that some nouns in English are much more likely to occur with the article the than otherwise (e.g., words of certain ontological classes, such as locative expressions, parts of inanimate objects, superlatives, and entities of nature, are statistically likely to occur with the in natural discourse). It concludes that when contextual knowledge becomes secondary, ontological knowledge and grammatical heuristics play a heavy role in the interpretation of noun phrases with the .