P1 Description

This project investigates how discourse particles in German (e.g. denn, schon, blo√ü) modify the illocutionary act carried out by a question, turning a simple information-seeking question like Wo wohnt er? “Where does he live?” into a non-canonical question (e.g. a rhetorical question). Combining tools from formal semantics/pragmatics and psycho/neurolinguistic research, it strives to identify and compositionally derive the syntactic, semantic and pragmatic constraints that govern this modification and to explore the cognitive mechanisms that underlie the processing and licensing of discourse particles in questions.

First Phase (2016-2019)

DiPs in interrogative environments are not restricted to the root clause, as shown in (i):

(i) Wo glaubt er (schon),dass man hier nachts um 3 Uhr (schon) t Benzin bekommt?
     where believes he DiP that one here at-night at three DiP gasoline gets
     “Where does he believe that one can get gasoline here at 3 o‚Äôclock in the night?”

(ii) Wer t glaubt (schon), dass man hier nachts um 3 Uhr (*schon) Benzin bekommt?
     who believes DiP that one here at-night at three DiP gasoline gets
     “Who believes that one can get gasoline here at 3 o‚Äôclock in the night?”

Three important observations have been made concerning these embedded occurrences: (A) even though embedded, the DiP modifies or “fine-tunes” the illocutionary type of the root clause; (B) the DiP must be located in the clause the wh-phrase has been extracted from or higher, witness (ii); and (C) the position of the DiP determines its propositional scope, yielding different readings for (i) depending on the position of the DiP. That is, modifying the illocutionary type at the root level and at the same time selecting a propositional argument at the embedded level can only succeed if the DiP lies somewhere along the wh-extraction path.

The first phase of the project has two goals. The first goal concerns the empirical description and psycholinguistic processing of examples like (i)-(ii). Empirically, it needs to be determined which embedding verbs show this pattern, which types of complement clauses allow for a embedded DiP, and whether pragmatic notions like “main point” have an impact on this construction. In terms of the human parser, we will investigate the role of locality during the integration of DiPs into the meaning of interrogatives and compare it to other semantic long-distance dependencies like Negative Polarity licensing. The second goal lies in semantics and pragmatics: modeling the meaning composition and final utterance meaning of examples like (i). Current analyses of DiPs are designed to account for fine-tuning of the illocutionary type of the local clause. The non-local pattern in (i) with an embedded DiP offers new ground to test competing theories of DiPs in combination with different semantic approaches to wh-phrases and with new pragmatic notions like main point status.

Second Phase (2019-2022)

Building on findings from the first phase of the project, the project’s goal for the second phase is articulated in three objectives.

First, while the first phase established correlates of various DiP-licensing violations, the second phase aims for a detailed assessment of the psycholinguistic processes underlying these correlates, and for their integration into the sentence comprehension literature. In two series of experiments, we will (1) disentangle the contribution of strictly linguistic processes from illusory licensing processes, and will (2) distinguish semantic and pragmatic processing pathways. All experiments will monitor the processing of DiPs and negative polarity items (NPIs) to allow a direct comparison and a better understanding of long-distance licensing processes at the interface of syntax, semantics and pragmatics.

Second, while the first phase concentrated on in situ discourse particles attached to the verbal spine, the second phase extends our investigations to ex situ particles forming a constituent with the wh-phrase. Pragmatically, ex situ but not in situ particles mandatorily carry an emphatic undertone. This raises interesting issues concerning the formal characterization of emphasis and its composition with wh-phrases. Semantically, ex situ particles have more scope freedom than in situ particles. Determining how the scope of the particles interacts with that of how many-phrases provides novel testing ground for current competing approaches.

Third, we are going to apply high-end methods for the analysis of neurolinguistic data to existing and new data collected in P1 and P6. While the first phase concentrated on traditional methods of neurolinguistic data analyses, the oscillatory brain dynamics in specific time-frequency bands will be systematically studied during the second phase to explore the cognitive mechanisms that underlie the processing of questions.