Tatjana Scheffler - Dissertation

Abstract

Semantic Operators in Different Dimensions

Tatjana Scheffler

Ph.D. dissertation, University of Pennsylvania, 2008.

Supervisor: Maribel Romero

This thesis studies the interface of truth-conditional and non-truth-conditional meaning by investigating constructions whose meaning and use differ because their semantic contributions are distributed differently over the semantic dimensions. The constructions in question are certain clausal adjuncts and complements.

For clausal adjuncts, I argue that two words for 'because' in German (weil and denn) contribute the same semantic operator (causality), but on different semantic dimensions. While weil operates in the assertion (or at issue) dimension, denn contributes instead a side comment (or conventional implicature). Consequently, the two words differ both in their range of use as well as in their semantic behavior as part of larger sentences. I point out the same empirical dichotomy for other adjuncts such as regular and relevance conditionals, although-clauses, and different kinds of adverbs. I show that for each of the constructions similar semantic differences result because an operator is contributed on the at issue dimension in one case, and as a conventional implicature in the other.

In the realm of complement clauses I investigate complements of attitude verbs. Of the large range of constructions that express the semantic arguments of attitude verbs, I study two in this thesis: slifting and embedded verb-second clauses. I show that these two constructions again mirror the situation as with weil and denn above: I propose that the two constructions contribute the same semantic pieces, but distribute them differently over the semantic dimensions of assertion and conventional implicature.

In multiple case studies, this thesis thus addresses some of the most important questions in linguistic semantics: What are the semantic pieces associated with a certain word or construction? How are these semantic pieces distributed over the known dimensions of meaning? And what effects does the individual distribution of meaning parts over semantic dimensions have for the overall meaning, function, and discourse effects of complex utterances?

The issue of the dimensionality of semantic entailments is not bound to a particular language (group), and the phenomena I study are generally cross-linguistically well-attested. For practical reasons, though, the discussion in this dissertation concentrates mostly on examples from German and English.