| What, from a semantic perspective, is the difference between singular indefinite and definite descriptions? Just over a century ago, Bertrand Russell provided a standard philosophical response. Descriptions are quantifier phrases, not referring expressions. As such, they differ with respect to the quantities they denote. Indefinite descriptions denote existential quantities; definite descriptions denote uniquely existential quantities. Now, while most philosophers were busying themselves defending and expanding upon Russell's position, some linguists were, quietly and independently, developing a radically different response. Descriptions, some linguists held, were referring expressions, not quantifier phrases. Accordingly, descriptions differ with respect to their rules of reference. Indefinite descriptions refer to 'novel' items, definite descriptions to 'familiar' ones.; Although some philosophers, most prominently Peter Strawson, would independently propose their own familiarity theories and although many linguists of a model-theoretic bent would join lock-step with the Russellians, little attempt was made to bridge the gap between these two seemingly incompatible responses. This dissertation composes the first systematic effort to bridge that gap. It provides a satisfactory answer to the above question which links the seemingly intractable divide between Russellians and familiarity theorists. This is achieved by utilizing two key observations: Donnellan's observation and Devitt's observation.; Keith Donnellan observed that speakers use descriptions not only as devices of quantification but also as devices of reference. Michael Devitt, following Donnellan, observed that these two uses, being regular, systematic, and cross-linguistic, have the status of convention in our language. Taken in conjunction, these two observations, I argue, require postulating that descriptions are semantically ambiguous. These observations compel the thoughtful theorist to maintain that descriptions have two distinct semantic functions, one quantificational and one referential.; Working within the confines of this ambiguity thesis, I argue that, in a certain sense, all parties to the dispute were right and wrong. On the one hand, Russellians provided a correct semantic account of quantificational descriptions. On the other hand, familiarity theorists presented a general framework that provides a correct account of referential descriptions. Accordingly, the semantic contrast between singular indefinite and definite descriptions is two-fold. Descriptions can contrast either quantificationally or with respect to the speaker's view of the audience's familiarity with the description's referent. Thus, by conjoining the previous responses' successes and disposing of their failures, I propose a theory that both successfully accounts for the semantic contrast of indefinite and definite descriptions and integrates two seemingly irreconcilable approaches. |
