An Examination of contriversy on "Metalibguistic Negation"

"Metalinguistic Negation" has been one of the main topics in pragmatics since Horn (1985). A great number of articles have been produced on this topic since then. This paper closely examines two distinctive opposite positions represented by Horn (1985, 1989, 2001^2) and Carston (1996, 1998, 2002), and tries to find what is the source of the difficulty in characterizing this phenomenon properly, which suggests another possibility of the unified semantic meaning of not. Horn (1985, 1989, 2001^2) distinguish the descriptive use of negation from the metalinguistic use of negation, based on whether the negation operator negates the truth-conditional content of the proposition which it operate on, and claims that negation is pragmatically ambiguous. Carston (1996, 1998, 2002) argue that the essential property of "metalinguistic negation" is that the material falling within the scope of the negation operator is to be understood as 'echoically used' (i.e. metarepresentation) and propose the new term "metarepresentational negation" for this phenomenon, and claim that the negation operator itself is standard descriptive truth-functional negation even in the cases of 'metalinguistic negation'. This paper closely examines these two positions and makes the points that Carston's metarepresentational negation covers virtually almost all the negation cases, if Givon's (1978) generalization is correct, that Horn's descriptive/ metalinguistic distinction roughly corresponds to her metaconceptual/metalinguistic distinction as subtypes of metarepresetational negation. It is further claimed that Carston's monogism on negation is convincing, though it is hard to keep her point as it is, considering the original logical definition of negation