Dr Hannah Sarvasy

Dr Hannah Sarvasy

Research focus: Hannah Sarvasy

29th January 2018

Hannah is one of three successful Australian Research Council Discovery Early Career Researcher Awards in CHL this year. She is a Postdoctoral Fellow in the ARC Centre of Excellence for the Dynamics of Language. Recently Hannah took the time to explain her very fascinating research project.

Millions of people around the world speak so-called 'clause chaining' languages. Speakers of these languages often use extra-long sentences that differ dramatically in structure--and likely require different cognitive processes to plan--than those of English. These extra-long sentences are called clause chains because they involve 'chains' of mini-sentences, or clauses. Despite being relatively easy to find in the world's languages, clause chains have had little impact on mainstream linguistic theory.

Indeed, for some time, much of linguistics has centred around short, English-like sentences. This approach overlooks both the fact that often we don't speak in complete sentences, but also that many languages have a different type of structure, where you can actually tell a whole story in one sentence, where in English the same story would have to be a whole series of sentences. In these languages you have a bunch of verbs ending in forms like the English 'ing': 'this morning waking up, brushing my teeth, doing this, seeing my friend, she coming in to tell me good morning, going to the store, buying something, going home.' It sounds like someone is listing something on and on. Then at the very end usually there is a fall in pitch, you have a tense marker (a form in language that establishes time in reference to the moment of speaking) that doesn't occur elsewhere in the sentence, as you would in English.

There are two things I want to look at. There are hundreds of clause chaining languages all around the world. Many languages of the Himalayas, a number of languages in Ethiopia, the Amazon and Mongolia, as well as Turkish and Japanese have these structures. So I want to look at them all and see what they have in common. No-one has ever done that in any depth.

The second thing is the idea of human processing capabilities. When we plan sentences in English we plan one or maybe a couple of words ahead. But in clause chaining languages there is a fascinating difference. I learned this when I was doing my PhD fieldwork in Papua New Guinea, in Nungon, a language that has clause chains.

When I am speaking Nungon it's clear to me that, despite speaking relatively fluently, I do not plan my sentences the way native speakers do. They flag switches in subjects ahead of time. For example, someone would say, 'I wake up in the morning, I brush my teeth, I see my friend,' and then use a special suffix that alerts the listener to an upcoming change: the next clause, or mini-sentence (verb and subject) is going to have my friend as the subject, not me. To be able to speak Nungon fluently, you have to know in advance what you're going to say, at least two mini-sentences at a time. But as a native English speaker I realised I wasn't planning enough; I was just spouting these little units one at a time.

I thus plan a two-pronged approach for the DECRA research. One is really a detailed survey of the languages, to see if their clause chains should really be considered similar structures, what ways they can vary, that kind of thing. The other is to study how people process clause chains, so starting in one or two languages and doing a couple of targeted experiments that may show that people are planning farther ahead in their language than has previously been thought possible, because we are basing so much theory on languages that don't have this structure.

The value of this research is that linguistic theory hasn't really accommodated what is a widespread feature of many languages throughout the world.

There are also implications for machine translation, so I want there to be more awareness of these forms. It will eventually make programs such as Google Translate more natural.

The most important thing for me is that I am proud to be able to go back to the Nungon community and others throughout the world and say 'it seems like you all plan your sentences ahead farther than I do. And that is something that has not been thought possible.' For small language communities something that simple can make a difference to how people think about their language and by extension their culture. These communities are often unsure if their language is something worth hanging onto when there is so much pressure to shift to another language, such as when the language of government and school is different.

To know that when their children speak Nungon their brains are stretching in ways unknown to speakers of more urban, prestigious languages such as Tok Pisin or English, will help the Nungon community to value their language for more than just its admittedly complex grammar. If their children are, in speaking their language, hardwiring their brains to be able to leap ahead of speakers of other languages then it helps to show that Nungon is something worth making the effort to preserve.

Updated:  7 July 2017/Responsible Officer:  Director, Culture, History & Language/Page Contact:  CHL webmaster