A corpus-based contrastive study tense, aspect, modality and polarity (TAMP) in Austronesian languages of Melanesia (MelaTAMP)
African, American and Oceania Studies
Applied Linguistics, Computational Linguistics
Final Report Abstract
This project has made a three-fold contribution to the state of the art in linguistics: Theoretically, methodologically, and in terms of primary data. Theoretically, we have gained new insight into the expression of time and reality status of events in Oceanic and beyond. Time and reality are intimately linked: What has already happened is undoubtedly real, whereas future events might always turn out other than expected. In talking about time and reality, languages differ in which of the two dimensions they prioritize. Oceanic languages, in particular those of Melanesia, are often described as prioritizing reality status over temporal reference, but prior to this project, this claim could not really be tested, because of a severe lack of data from the relevant languages. We have accessed language-documentation data from seven Oceanic languages of Melanesia to investigate their ways of expressing temporal reference and reality status. Based on preliminary empirical observations, we developed a new theoretical approach within the tradition of temporal/modal logic and semantics to articulate very specific hypotheses and research questions. We further refined these on the basis of our data collections from language documentation. We then designed new stimuli for testing in the field. During a highly concerted effort, five international researchers collected new data with our stimuli in six of our subject languages, in remote locations in Vanuatu. Our results show that our subject languages vary significantly in whether they prioritize reality status over temporal reference. Our theoretical approach also allows us to model the rather gradient differences between languages that we found. We also did the first comparative survey of the expression of habitual aspect in this group of languages, that is, regularly recurring situations as in I go to college. We further investigated the relation between perfect aspect (I have been to Paris) and already and gained new insight into how the overlapping function of these expressions vary across languages. Methodologically, we wanted to highlight the vast potential of corpora from language documentation for comparative studies. This is one of the first project that has successfully tapped into the promise of this novel resource, which is still very hard to access and explore, especially for researchers who are not part of the language-documentation community. Our usage of these data was greatly facilitated by our research software engineer, who has also worked on the topic of sustainability of research software. In terms of primary data, this project has produced substantial amounts of new recordings, which have been fully transcribed, translated, and partially glossed. These include data from moribund languages such as Mavea, which only has an estimated 30 speakers left. Working with picture stories has proven to offer several significant advantages, in that it creates highly targeted, but still semi-spontaneous speech, semi-parallel corpus data for the investigation of language-internal and cross-linguistic variation, and the results can be used for the production of annotated picture books for speaker communities.
Publications
- (2016): Realis and non-realis modalities in Daakie (Ambrym, Vanuatu). Proceedings of SALT 26, 566-583
Krifka, Manfred
(See online at https://doi.org/10.3765/salt.v26i0.3865) - (2018): A comparative study of conditional clauses in Nafsan. Proceedings of COOL 10
Krajinović, Ana
- (2018): Mapping Irreality: Storyboards for eliciting TAM contexts. Proceedings of Linguistic Evidence 2018
von Prince, Kilu, Manfred Krifka, Ana Krajinović, Valérie Guérin and Michael Franjieh
(See online at https://doi.org/10.15496/publikation-32623) - (2019): A software-driven workflow for the reuse of language documentation data in typological studies. Proceedings of the Workshop on Computational Methods for Endangered Languages
Stephan Druskat and Kilu von Prince
(See online at https://doi.org/10.33011/computel.v2i.453) - (2019): Habituality in four Oceanic languages of Melanesia, in: Nora Boneh and Łukasz Jędrzejowski (ed.). Special issue of STUF. 72(1)
von Prince, Kilu, Ana Krajinovic, Anna Margetts, Valérie Guérin, Nick Thieberger
(See online at https://doi.org/10.1515/stuf-2019-0002) - 2019. Counterfactuality and Past. Linguistics and Philosophy
von Prince, Kilu
(See online at https://doi.org/10.1007/s10988-019-09259-6) - 2019. Expressing possibility in Daakaka and Saliba- Logea. Studies in Language
von Prince, Kilu and Anna Margetts
(See online at https://doi.org/10.1075/sl.18051.pri) - 2019. Habituality as a property of text spans, in: Zygmunt Vetulani and Patrick Paroubek (eds.) Human language technologies as a challenge for computer science and linguistics— 2019. Poznán: Wydawnictwo Nauka i Innowacje, 195-199
Tjuka, Annika, Lena Weißmann, Kilu von Prince
- 2019. The semantics of perfect in Nafsan and implications for typology. Proceedings of TripleA 5 2018
Krajinović, Ana
(See online at https://doi.org/10.15496/publikation-36879)
