O teście negacji, kontekstach interpretacyjnych i zdaniach „wyratowanych”*
On Negation Tests, Interpretational Contexts and „Saved” Sentences
Languages of publication
In the article I address several problems concerning the role of so called negative linguistic evidence in semantic analysis. The focus is on two main questions. The first addresses the problem whether sentences inherently contradictory, used as justifications for semantic theses, should be anomalous in an absolute sense (i.e. whether they should remain incoherent in any thinkable context), or whether it is possible to “save” them in some contexts, without losing their value as semantic evidence. I point out that, if one aims at demonstrating contradiction on the surface of a sentence (by confronting a given lexeme with the negation of its alleged semantic components), one should not expect any irrefutable proof. On the other hand, grammatical anomalies can serve as excellent evidence in semantics, but sentences of this kind are not always possible to apply, and after all, they only give indirect evidence without making a semantic component in question explicit. If a contradiction concerns only the lexical level of a test sentence, it is possible that one will always be able to “save” the sentence; still, it does n o t mean that it is not contradicted, yet if we are to grasp this contradiction, we must apply strict criteria of sentence “codability”; I assume (after Bogusławski 2008, 2009 and Frege 1977 ) that they are supplied by the frame of the predicate somebody said that_ together with the assumption that a statement in question was made seriously. The second issue addressed in the article focuses on how, exactly, a contradiction should be presented in a test sentence – since different grammatical constructions or different metatextual operators can render different test results. The analysis of some sentences referred to in several semantic papers shows that it is not a good idea to use conjunctive constructions with i ‘and’: although such sentences could easily be marked with an “alienating asterisk”, it is often not because they are inherently contradictory, but because of high “coherence requirements” characteristic of the conjunction i that are not fulfilled in them. For this reason it happens that the very cause of anomaly is interpreted incorrectly.
Publication order reference