Morphological marking of topic-comment relations in East Asian languages

The fundamental guiding force of human conversation is the “Question Under Discussion”. Implicitly, when we have a conversation we set a QUD or “topic” and then we add new information as the focus or comment. We keep adding new information until the QUD is resolved or we branch out to a new topic. Learning how to manage discourse structure in this way takes considerable developmental time. Children have to learn how to make it clear to their listener what their topic is, what information they are adding and how this new information relates to the topic. Different languages provide different tools for achieving this. Germanic languages like German, Dutch and English rely heavily on prosody – we change the pitch, amplitude and duration of speech to emphasise words in various ways to make their contribution to the discourse clear (Chen, 2017). It is difficult to study the development of topic-focus relations in such languages since it requires time-consuming analysis of spectrograms and pitch contours in the absence of clear cut coding criteria. Other languages, however, mark topic-focus with special pragmatic morphemes just for the job. For example, in Japanese speakers can add the particles ‘wa’ and ‘ga’ after nouns to mark something akin to topic and focus respectively. Similar particles exist in Korean and other related languages. Indeed, in contrast to European languages, East Asian languages generally tend to organise their grammars far more around the pragmatic notions of topic and focus (how what is being said relates to the question under discussion ) than around semantic notions of agent and patient (who is doing what to whom). A handful of studies to date suggest that children are quite quick to start using topic and focus particles in these languages but are very slow to become fully adult-like in their use. This project could assess how development proceeds by looking at corpora of parent-child conversation available in the CHILDES talkbank. An example corpus study (Kurumada, 2009) is available below.
For this project, fluency in either Japanese, Korean or any language with topic/focus particles and freely available corpora is necessary.