Servicenavigation

Project Details

Back

Projekt Print View

Modelling Control Theory

Subject Area General and Comparative Linguistics, Experimental Linguistics, Typology, Non-European Languages
Term from 2014 to 2018
Project identifier Deutsche Forschungsgemeinschaft (DFG) - Project number 258946513
 
Final Report Year 2018

Final Report Abstract

In this project, we set out to explore the syntactic treatment of control structures. Our initial objectives were the following: conceptually, we aimed to develop a comprehensive syntactic theory of control based on the phase model as underlying framework; empirically, our starting points of investigation were control into islands and backward control (BC). We could show that (i) obligatory control (OC) into islands poses a severe problem for the Movement Theory of Control (MTC), and that (ii) BC is no argument for the MTC, in contrast to what is standardly assumed. As far as control into islands is concerned, we did not only investigate control into adjoined islands (as it has been done in the literature before), but showed moreover that OC into non-adjoined islands is a frequent pattern in German, for which the MTC cannot account. As an alternative, we advanced the Hybrid Theory of Control (HTC), which assumes that OC is licensed under upward Agree between the controller and the controllee with the Phase Impenetrability Condition imposing locality restrictions on the possible distance between probe and goal. As regards BC, we investigated Greek, Romanian, and Spanish, and argued that the Greek/Romanian type reduces to restructuring, while the Spanish type relies on 9-feature sharing under Agree. Crucially, BC does not involve movement and therefore does not provide an argument for the MTC. The analysis we put forward also accounts for the fact that backward subject control is fully productive in Greek, whereas backward object control is severely restricted. Our work on control into islands led moreover to a follow-up study on control into subject clauses of object experiencer verbs. If they are not embedded, they display an OC reading; if they are embedded, long distance (LD) control can occur, which signals non-obligatory control (NOC). Contrary to what has been claimed in the literature, we showed that extraposition does not have an effect on this NOC reading. We suggested that semantics comes into play in the case of LD control insofar as it involves the attitude holder as controller, which we argued to be projected in the syntax. The OC/NOC distinction also played a role in our analysis of adjunct control. Although the OC/NOC distinction cuts through the set of examples involving adjunct control, we argued that there is a correlation between the height of the adjunction site and the type of control that is established. This strongly argues for a syntactic treatment of control, and we showed that the HTC can account for a large set of data from German, Norwegian, and English. With respect to partial control (PC), we provided the first large-scale experimental investigation of PC in German and got the following results: the acceptability of sentences involving PC predicates that do not allow comitatives suggests that "true PC" occurs in German, meaning that the PC analysis in these cases cannot rely on a comitative-based account. On the other hand, however, non-PC predicates (i.e. exhaustive control predicates) that allow comitatives also give rise to a PC reading ("fake PC"), which suggests that there must exist two ways to derive a PC reading in German. For implicit control, we argued that the following correlation holds across languages: implicit predicative control can only occur in languages which allow impersonal passives, because the former must be construed as an impersonal passive; implicit logophoric control, by contrast, may be construed as a personal passive via an anaphoric pronoun it. This option is not available in the case of implicit predicative control for semantic reasons. We also had a close look at Polish control with an emphasis on constructions involving an embedded Case-marked adjective, and developed an account which can derive the observed Case patterns on the basis of different underlying predication relations. To sum up, we investigated control into islands, control into subject clauses of object experiencer verbs, adjunct control, partial control, implicit control, subject and object backward control, Case and control, and the distinction between OC and NOC. Although we pursued a local-derivational syntax-based approach to control, we came to the conclusion that PC and implicit control readings are presumably rooted in the semantics, and that semantic factors also play a role in attitudinal contexts, in which semantic and pragmatic factors determine the attitude holder which we argued to be syntactically encoded allowing for a syntactic treatment of logophoric control.

Publications

 
 

Additional Information

© 2024 DFG Disclaimer / Copyright / Privacy Policy

Textvergrößerung und Kontrastanpassung