Christian Fortmann

Syntactic aspects of the SLF-coordination (subject gap in finite clause coordination) in German are investigated. Some empirical observations give rise to a revision of the conditions that license this type of construction: i) SLF-coordination is subject to a constraint on constituent order: scrambling in the first conjunct is restricted if the predicates in both conjuncts are formed by transitive verbs; ii) adverb-placement in the second conjunct is impossible if the adverb must be inserted in a position preceding the gap; iii) non-subject-gaps are possible with a subclass of verbs if the gap precedes the subject in base word order. These empirical facts are incompatible with analyses that either exclude the subject from the coordinated phrases or restrict the gap to the subject function. In this contribution it is argued that both conjuncts have to be parallel with respect to their respective focus potential and that the gap is restricted to a position immediately following the finite verb. In order to capture the latter fact, a condition on NP-ellipsis is proposed building on Phillips (2003) theory of incremental structure derivation.

Ingo Reich / Marga Reis / Veronika Ehrich / Christian Fortmann: Einführung.

Katrin Axel: Die Entstehung des dass-Satzes - ein neues Szenario.

Ulrike Demske: Zur Markierung von K [...]

The interrogative warum connected with a bare infinitival VP forms a specific kind of root structures in German. wh-root-infinitives in general are subject to certain syntactic and interpretative restrictions. The subject argument of the infinitival verb cannot be expressed by a lexical NP; it is identified with the speaker, instead. Due to the lack of finiteness these infinitives are restricted to a non-past, prospective, modal interpretation. warum-infinitives share these properties. But differing from other instances of wh-infinitives they are systematically ambiguous, as they provide a second reading with a non-prospective interpretation referring to a present or past event in which the subject argument is identified with the addressee. It will be argued that the availability of this second reading has to be ascribed to a pragmatic inference which refers to presuppositions of questions in general and of causal questions in particular.