Periodic Reporting for period 1 - AAR-ATB (Argument asymmetries revisited: Variation in parallelism constraints on ATB-movement)
Reporting period: 2023-10-01 to 2025-09-30
A long-standing question in syntactic theory pertains to how to account for the different status of subjects and objects with respect to grammatical processes. Several decades of research in syntax have uncovered numerous ways in which subjects behave differently from objects. For example, in English and several other languages, objects have been argued to be transparent for sub-extraction, whereas subjects are typically not. A key question is what such asymmetries between arguments can tell us about the nature of grammar. What are the grammatical primitives that distinguish subjects from objects, for example? Must the relevant constraints make reference to grammatical function, or are such restrictions the result of the differing structural prominence of arguments? Answering questions of this kind can reveal deep facts about the building blocks that underpin our implicit knowledge of grammar. What abstract principles does a grammar require in order to adequately account for argument asymmetries?
This project aimed to investigate the phenomenon of parallelism conditions in so-called Across-the-Board dependencies. On the empirical side, the project aims to broaden the empirical base through a comparative study of tolerable mismatches between argument types in ATB-constructions on the basis of three languages (English, German, Polish) that also takes into account the role of variation. On the theoretical side, the project will incorporate these findings into a general theory of cross-linguistic argument asymmetries involving movement, a goal which has not yet been seriously pursued in contemporary syntactic research The general goals of the project can be divided into both an empirical and a theoretical contribution. On the empirical side, the project will collect data based on native speaker judgements from a range of speakers of English, German and Polish. The surveys will be designed in such a way as to provide a systematic empirical base to test both the predictions of certain theories of ATB-parallelism and also the degree of variation we find across speakers and languages. This will provide a solid empirical foundation on which theoretical work can be conducted. It will also provide a clearer picture about the central explananda for a theory of ATB-movement that more carefully takes into account the potential role of variation. With this improved empirical base, the theoretical contribution of the project will be to integrate the findings into a maximally general theory of argument asymmetries. As previously mentioned, the pursuit of a unified approach to argument asymmetries has been abandoned for entirely orthogonal reasons. There are, however, new concepts that can fill this theoretical void (i.e. phase theory) and provide a general explanation of the recurrent matrix subject vs. non-subject asymmetry we find across languages and domains. The project takes ATB-parallelism as a detailed case study in one such domain, but will ultimately seek to situate these findings in the relevant cross-linguistic context. Doing so, will allow us to arrive at a deeper understanding about what the cross-linguistic landscape of argument asymmetries tells us about the nature of syntactic knowledge and our capacity for language.
This project aimed to investigate the phenomenon of parallelism conditions in so-called Across-the-Board dependencies. On the empirical side, the project aims to broaden the empirical base through a comparative study of tolerable mismatches between argument types in ATB-constructions on the basis of three languages (English, German, Polish) that also takes into account the role of variation. On the theoretical side, the project will incorporate these findings into a general theory of cross-linguistic argument asymmetries involving movement, a goal which has not yet been seriously pursued in contemporary syntactic research The general goals of the project can be divided into both an empirical and a theoretical contribution. On the empirical side, the project will collect data based on native speaker judgements from a range of speakers of English, German and Polish. The surveys will be designed in such a way as to provide a systematic empirical base to test both the predictions of certain theories of ATB-parallelism and also the degree of variation we find across speakers and languages. This will provide a solid empirical foundation on which theoretical work can be conducted. It will also provide a clearer picture about the central explananda for a theory of ATB-movement that more carefully takes into account the potential role of variation. With this improved empirical base, the theoretical contribution of the project will be to integrate the findings into a maximally general theory of argument asymmetries. As previously mentioned, the pursuit of a unified approach to argument asymmetries has been abandoned for entirely orthogonal reasons. There are, however, new concepts that can fill this theoretical void (i.e. phase theory) and provide a general explanation of the recurrent matrix subject vs. non-subject asymmetry we find across languages and domains. The project takes ATB-parallelism as a detailed case study in one such domain, but will ultimately seek to situate these findings in the relevant cross-linguistic context. Doing so, will allow us to arrive at a deeper understanding about what the cross-linguistic landscape of argument asymmetries tells us about the nature of syntactic knowledge and our capacity for language.
The commencement of work on the first work package was delayed by approximately 3-6 months due to the emergence of additional relevant projects and also certain complications surrounding the construction of experimental stimuli. Additionally, it became necessary for there to be an extended period of literature consultation in order to adequately create naturalistic stimuli for both Polish and German. Here, the choice of both verb type and filler for the ATB-extraction contexts required further consultation with native speaker than had been initially anticipated. For this reason, the experimental design and data collection was not fully completed by the end of the initial 12 month period envisaged in the work programme. Of the four studies outlined in the first work package, stimuli were created for three studies. These were constructed (together with fillers) and were ready to be coded in Qualtrics, before the premature conclusion of the project. The materials for two further studies were not able to completed, as it became apparent that it was difficult to find enough different ditransitive and dative-assigning transitive verbs that would create relatively naturally-sounding stimuli. Here, it was decided that one would first require some pilot investigation into the general acceptability of such conjunctions, before it could be justified to the run them on a larger scale.
There was significant progress made toward a better understanding of argument asymmetries and their representation in grammar. Worthy of mention here is one joint sub-project in particular. The topic of this project was to empirically test the predictions of a particular theory of ‘Across-the-Board movement’ with relation to whether there is any empirical evidence for the syntactic presence of non-adjacent gaps. The results of the project was accepted to a prestigious international conference. There is now also a proceedings paper from the conference, which will appear in the near future.
Although the premature termination of the project meant that the impact of this objective was also somewhat less than had been anticipated, there was some significant progress made toward contributing to the theoretical understanding of argument asymmetries. The project also involved work on a collaborative project on agreement asymmetries in the Mundari language. This project deals with asymmetries with verbal agreement in Mundari, whereby the verb sometimes agrees with the direct object and sometimes with the indirect object. Illuminating the conditions behind this variable pattern, which was shown to be subject to the various preference hierarchies at play in the language, made an important contribution to our understanding of the way languages may variably privilege one argument type over the other (and how this may be linked to other factors). After the conclusion of the fellowship, resulting paper has now been published in a top-tier journal.
Although the premature termination of the project meant that the impact of this objective was also somewhat less than had been anticipated, there was some significant progress made toward contributing to the theoretical understanding of argument asymmetries. The project also involved work on a collaborative project on agreement asymmetries in the Mundari language. This project deals with asymmetries with verbal agreement in Mundari, whereby the verb sometimes agrees with the direct object and sometimes with the indirect object. Illuminating the conditions behind this variable pattern, which was shown to be subject to the various preference hierarchies at play in the language, made an important contribution to our understanding of the way languages may variably privilege one argument type over the other (and how this may be linked to other factors). After the conclusion of the fellowship, resulting paper has now been published in a top-tier journal.