This paper is about German finden as a subjective attitude verb (Sue findet, dass das gut ist/ Sue findet das gut), about which I aim to find out in what sense it has 'subjective meaning', and how this meaning is rooted in its lexial properties. Based on an intensive discussion of Saebø's (2009) influential analysis, I argue, contra Saebø, that finden's subjectivity is not exclusively rooted in the complement let alone identifiable with semantic judge-variance of the complement predicates. Rather, finden itself is crucially involved in that it projects (i) the presupposition that there is an eventuality known to the finden subject, which is under debate and open to interpretation, (ii) a reading of the complement proposition as an interpretation of this eventuality selected by the finden subject from a set of interpretational alternatives; it is with this selection process that the finden specific subjectivity is to be identified. I cite a number of arguments for these claims (summed up in (H1)), backing them up by pragmatic considerations (summed up in (H2)). Finally, I show that the syntactic form of the finden complement as a finite vs. small clause construction correlates with selectional differences, for which I offer a speculative explanation.

Typically, the semantics of concessive conjunctions such as obwohl in German or although in English has been described by taking recourse to conditionality. König (1991b), for example, claims that obwohl has two meaning parts: (i) It has the truth-conditional meaning of the logical conjunction ∧ and (ii) it expresses the concessive presupposition p → ¬q. In this paper, we argue that the underlying conditional semantics of obwohl has to be analyzed as a tripartite quantified structure with an implicit modal in the sense of Kratzer (1978, 1991, 2012). This analysis has the advantage that the two meaning components of obwohl do not enter in a logical contradiction as in the traditional analysis. Furthermore, the specific semantic and pragmatic properties of German spoken variants of concessive clauses (i.e. obwohl-clauses with V2-order and disintegrated obwohl-clauses with verb-final order) can be derived without further assumptions within the quantificational analysis proposed in this paper.

This paper will propose that the various patterns of clitic doubling across Romance (and other) languages are the result of diachronic developments, namely of the interaction of catastrophic and gradual processes. More precisely we will argue that catastrophic processes (the emergence of doubling) need to be differentiated from gradual processes (the spread of doubling to different contexts). Under this view the catastrophic change is the change concerning one macro parameter giving way to several micro parameters which are then responsible for the spread of clitic doubling to different syntactic contexts.

This article proposes that the interpretation of sentence mood in German discourse is essentially controlled by the use of different intonation contours. We claim that [-wh]-sentences (declarative questions, progredient declaratives), V1-sentences (polarity questions, conditionals) and [+wh]-sentences ([+wh]-questions, echo-questions) are distinguished not only by rising vs. falling tonal movements but by different types of rising contours (concave vs. convex tonal movements). These contours also discriminate types of the discourse particle hm and determine their interpretation, although in an opposite distribution than with the sentences. To complete the picture of the expression of sentence mood, V1-declaratives and exclamatives are integrated into the discussion. This study opens up new perspectives to sentence mood theory in that it considers intonation as a non-truth-functional meaning forming part of sentence interpretation.