Pragmatics is already an established subfield of linguistics and the Gricean distinction between literal and speaker's meaning is one of the foundations of modern linguistic theory. But progress in pragmatics has been slow compared to other subfields of linguistics over the last four decades. We argue that one recent trend, namely Experimental Pragmatics, promises to overcome the stagnant state of pragmatic theory. We present both the three main developments that come together in Experimental Pragmatics (Gricean pragmatics, precise models, and formal experiments). We then present recent results that exemplify the promise of Experimental Pragmatics in two core domains: scalar implicatures and metonymy.

In calculating the temporal order of events introduced in discourse at least two main factors have been recognized to be relevant beyond the linear order: potential causal links between the events and aspect/tense marking on the involved verbs. While causality has been argued to play an important role in general theories of discourse structure such as Segmented Discourse Representation Theory (SDRT, Asher and Lascarides 2003), the effect of aspect/tense marking has often been discussed in theoretical approaches to tense and aspect, e.g. Kamp and Rohrer (1983) or Klein (1994). The most complete theory of the functions of aspect in the computation of temporal relations is the theory of Dynamic Aspect Trees in ter Meulen (1995). Typically, both kinds of approaches tend to ignore the importance of the particular other factor. In this paper we present the results of an empirical study that investigates the strength of the respective factors in interpreting German two-sentence discourses and the impact they have on the acceptability of such discourses. We show that both factors play a significant role, thus none of the theoretical approaches can prevail without the other. However, we also found that the effect of causality is stronger than the effect of temporal marking in some cases, which leads us to suggest that an integration of the insights of ter Meulen (1995) into SDRT is required.

This paper introduces the idea of graphematic word formation. For this purpose, I will discuss the criteria of morphological word formation with respect to abbreviations ending in an abbreviation mark: means ,abbreviation' or means ,for example'. The criteria are: 1. There is a semantic shift between the original unit and the newly formed unit. 2. The newly formed unit may develop its own inflection. 3. The newly formed unit may belong to a different part of speech as the original unit. 4. The newly formed unit may consist of direct and binary constituents. 5. The newly formed unit may be linked to a proper pronunciation. Abbreviations fulfill the criteria, which are necessary for morphological word formation but they do not develop a proper pronunciation. An abbreviation's semantics can differ from the semantics of the original longer term and abbreviations can develop its own plural inflection. The paper's second part deals with the graphematic structure of abbreviations. Abbreviations ending with one abbreviation mark consist of three to five letters on average. They build graphematic syllables like any other graphematic word. Internal majuscules can furthermore occur in abbreviations as well as in acronyms to structure the graphematic word. The divis is suggested to function as a graphematic linking element.

In the literature, parentheticals have often been claimed to be independent of their anchor clause. This article discusses the syntactic independence of German declarative V2-parentheticals and the question of whether they are prosodically set off from their anchor clauses. For parentheticals in English, it has been stated that prosodic features may be neglected (see Bolinger 1989; Wichmann 2001). Based on corpus data of German V2-parentheticals, the result will be replicated. This contribution to the research on parentheticals is mainly empirical and limited in scope. The results are the following. The claim that V2-parentheticals are syntactically independent of their anchor clauses can be supported. However, the claim that V2-parentheticals are quieter, faster, and lower than their anchor clause and framed by silent pauses is argued to be untenable. Even a weakened version of the claim must be rejected. A functional explanation for why prosodic independence is not a necessary condition for V2-parentheticals (as, perhaps, opposed to other parenthetical constructions) may lie in the fact that V2-parentheticals are already marked syntactically, namely by means of their V2-structure inserted in non-V2-places (in the middle field, within NPs/DPs).