SCOPIC widens its scope
The cross-linguistic semantic typology project Social Cognition Parallax Interview Corpus (SCOPIC) has reached its midway point, ramping up work to resolve the most wicked problems thrown up by this world-first initiative.
The project is producing an annotated corpus of (currently) 27 languages from every continent of the world, representing a wide spectrum of cultures. The corpus is being used to study how social cognition is represented and managed by speech communities. It gathers enriched language data through an innovative, structured task that involves collaborative, narrative problem-solving and retelling by a pair or small group of speakers responding to picture cards.
The pictures set up a dramatic story – ‘Family Problems’ – where participants can feel empathetic involvement with the characters and trace individual motivations, mental and physical states, and points of view. This form of data-collection allows different cultural groups to imbue the pictures with their own experiences, concerns and conventions, and stimulates the spontaneous use of previously under-recorded linguistic structures.
In 2018, SCOPIC members held week-long workshops in Cologne (Germany) and Tokyo, leading to successful new grant proposals. Two collaborators, Henrik Bergqvist of the University of Stockholm, and Asako Shiohara of the LingDy project at the Tokyo University of Foreign Studies (TUFS), obtained funding to support further workshops during 2019 and beyond.
SCOPIC member Elena Skribnik and CoEDL-based project leaders Nick Evans and Danielle Barth also co-taught a course for students at the International Winter School of Finno-Ugric Studies in Vienna.
L to R: Keita Kurabe, Yukinori Kimoto and Norikazu Kogura on their visit to Australia.
Three Japanese postdocs from LingDy – Norikazu Kogura, Yuknori Kimoto and Keita Kurabe – spent time at CoEDL under the Japanese 'Brain Circulation' scheme, which included intense participation in the Australian SCOPIC workshop. Additionally, Kogura was able to work with the Sibe language, a variety of the severely endangered spoken Manchu language, which has a significant diasporic population in Australia.
While in Australia, Kogura and Kimoto had a special week focused on understanding the connection between complementation and reported speech with ANU project members. The researchers are comparing nine meanings that are often expressed with complement clauses; e.g. perception, fear or attitude (“I think, believe, see… that she is a good person.”).
However, not all languages have complement clauses. Reported speech, which refers to how a speaker talks about how other people talk, think and feel, is another commonly used way of expressing similar meanings in other languages; e.g. “I thought this: ‘she is a good person’”.
So the researchers face the challenge of coding the data in the corpus in such a way that natural conversations from 27 very diverse languages are scientifically comparable. And when we compare languages that do not have complement clauses, how to they transmit the same social meanings? Is reported speech the most common way to do this, especially when the embedding is multi-layered?
In March, project members based in Japan and Indonesia had a successful meeting with Nick and Danielle at TUFS, hosted by Shiohara. Here, Kogura and Kimoto presented a new schema for coding complementation and other syntactic strategies for presenting commentary of propositions. The next project meeting, hosted by Bergqvist and held in Stockholm the following week, carried this work further by testing the schema with new languages, expanding, adapting and finalising the work.
The SCOPIC team hopes that coding this cross-linguistically will allow them to investigate and test hypotheses about the origins of structural complexity in language.
Participants at the SCOPIC meeting in Stockholm, April 2019.
(Main image: Norikazu Kogura, Danielle Barth, Honoré Watanabe and Nick Evans at the SCOPIC meetings in Tokyo, March 2019)