Seminar Series

Department of Linguistics seminars are held on Fridays, 12:00–1:30pm.
Enquires: ,

24 February: Workshop organised by Maïa Ponsonnet
Rogers Room, Woolley building

Emotional ‘sprinklers’: Interjections and evaluative morphology in Australian languages and elsewhere
Download the workshop program, PDF (150KB)

17 March: Anna Vatanen, University of Helsinki
Rogers Room, Woolley building

Anna is visiting our Department at The University of Sydney this semester.

Silent moments in Finnish everyday interactions: co-presence, conversational activities and participant orientations

In this talk I will explore moments of silence in interaction, especially the lapses that occur between sequences of talk. These silences can be characterized, at least potentially, as “the conspicuous absence of talk” (see Hoey 2015). Data from videotaped, naturally occurring everyday interactions among intimates will be examined with conversation analytic methods. Participants in the data are speakers of Finnish, who are stereotypically regarded as favouring silence in situations where speakers of many other languages assumedly prefer talk. I will examine what happens both before, during and after the lapses.

First, during the lapses the participants use certain embodied and vocal means to manage the lapse and to display their orientation to it as possibly problematic; here discussion on co-presence and (dis)engagement become relevant. Second, exploring pre- and post-lapse talk will bring in the issues of conversational activities and topic talk: it seems that there is a relationship between the ongoing activity and the participants’ observable orientation to the lapse. Furthermore, I will show that the participants exploit specific linguistic resources for indicating whether their post-lapse talk is expanding and tying to the topic prior to the lapse or whether they are initiating something new.

With these observations I hope to provide some evidence that real-life Finnish interactions match the widely held stereotype at most only marginally/partially.

31 March: Mark Ellison, The Australian National University
S226 (MECO), Woolley building

From Speakers' Cognitive Biases to Macrolinguistic Diversity

Some parts of the world are rich in diverse languages, while others are linguistically uniform. For example, Samoa and Vanuatu are comparable landmasses in the South Pacific, with similar populations and settlement periods. However, Samoans speak a single language, while on Vanuatu more than 100 are spoken.

In the project The Wellsprings of Language Diversity at ANU, we are exploring potential accounts for these disparities in diversity. One focus of the research studies the relationship of within-group variation across a range of language communities (in Australia, PNG, Samoa and Vanuatu), with the goal of correlating this microvariation to macrodiversity over longer time scales. Part of building robust inferences from microvariation with macrodiversity is computational modelling, thus this is forms an important part of the project.

In this seminar, I will present my work both within and without the project which bears on the micro causes of language diversity. An important part of this work focusses on speaker biases - such as egocentric and frequency biases, as well as the anti-doppel bias in bilinguals - which over time can act as forces favouring linguistic uniformity or diversity. These biases underly the speaker simulations in "close" models - simulations of (necessarily small) linguistic communities at the level of identified individual speakers. In one case, we are constructing a model of a southern PNG village (Bevdvn) incorporating all living individuals, their home and gardens, and their interaction patterns. With such a detailed model, we can explore how much cognitive biases can explain the progress of observed linguistic change in the community, leading to sychronic diversity and eventually linguistic change.

28 April: Gisela Håkansson, Lund University, Østfold University college (Norway) and Linneaus University (Sweden)
S226 (MECO), Woolley building

Psycholinguistic and typological aspects of language development

In my talk I present research on language acquisition from the perspective of Processability Theory (PT: Pienemann 1998, 2005, 2015 etc.). PT is a psycholinguistic account of the developmental sequences found in the acquisition of morpho-syntax. The proposal is that processing procedures activate the emergence of L2 morpho-syntactic structures in a given order. Drawing on a large body of work on Swedish and other Scandinavian languages I will give examples from L1 and L2 development and also present data from children with language impairment learning those languages.

Another perspective is the typological validity of PT. Up until now, the L2 acquisition of 17 different languages has been documented, not only Indo-European languages but also languages from five other language families (Afro-Asiatic, Altaic, Austronesian, Japanese and Sino-Tibetan). I will present some of the structures that emerge early and late in the L2 development of different languages.

19 May: Gwendolyn Hyslop, The University of Sydney
Rogers Room, Woolley building

Tonogenesis in Kurtöp

According to the established model of tonogenesis (e.g. Haurdicourt 1954; Matisoff 1970, etc.), a contrast in consonant type conditions pitch on the following vowel. Once pitch has phonologized, the original contrast in voicing ceases to be distinctive. Kurtöp, a Tibeto-Burman language of Bhutan, is in the process of undergoing tonogenesis and thus provides an opportunity to study the sound change in detail. This talk presents the results of a production study, showing how the initial contrast in voicing of obstruents is slowly being replaced by a contrast in pitch on the following vowels. In particular, we see that high pitch has phonologised following the voiceless initials while low pitch has phonologised following the voiced initials. The voiced initials are now in the process of devoicing. However, this process is not happening at equal rates across the different manners and places of articulation. Specifically, we see that fricatives are more likely to be devoiced than stops, and within the domain of stops, we see that retroflexes are most likely to be devoiced while bilabials are least likely to be devoiced. These results taken together, and in the context of tonogenesis in a typological perspective, suggest that, as a sound change, tonogenesis can be shown to actually follow a predictable pathway, making use of acoustic universals.

2 June: William Foley, The University of Sydney
Rogers Room, Woolley building

A typology of nominalizations: a Role and Reference Grammar approach

Syntactic nominalizations with their hybrid, mixed category properties have long presented a challenge to grammatical theories and attracted much theoretical interest (syntactic verbalizations should too, but they have almost been completely ignored). A syntactic nominalization is a (typically morphological) derivation which produces a phrase, that is, when a verb and its associated core and, potentially, peripheral arguments and modifiers are converted into a noun phrase. A verbalization would be the opposite, when a noun phrase with a head noun and any associated arguments and modifiers is converted into a verbal constituent. Crucially these both involve a morphological derivation on the head word, but result in a constituent or a dependency larger than a word. Such syntactic nominalizations are extremely common, if not universal, but true syntactic verbalizations are quite rare, to my knowledge only attested clearly in Eskimoan (Sadock 1980, 1985, 1986) Wakashan (Stonham 2008; Waldie 2004) and Munda (Peterson 2011, 2013) languages, where the process is called noun incorporation or denominal verbalization. Why there is such an asymmetrical distribution of these two processes is an interesting question in its own right, but will not be pursued here, which will focus on nominalizations. The mixed categorical, hybrid nature of nominalizations, realized in the fact that they often exhibit simultaneously the properties of both basic underived nouns and verbs, has been treated in a number of different ways, for example, the dual feature specification [+N, +V] in Lefebvre and Muysken (1988), (though this will not work crosslinguistically, since it fails to distinguish nominalizations from adjectives), I adopt here the proposal of a dual headed structure with both noun and verb heads, each with their own projections, of Bresnan (1997). However, Bresnan’s phrase structure based framework has limitations in treating nominalizations fully in their crosslinguistic typological variation because it lacks a theory of scope and operators which are crucial to describe the various levels at which nominalizations can occur. Many languages including English have varying levels of nominalizations with different mixes of nominal and verbal properties like Leonardo painting the Mona Lisa, Leonardo’s painting the Mona Lisa, Leonardo’s painting of the Mona Lisa, Leonardo’s paintings, and analyzing this requires a rich layered theory of constituent structure, each with its own operators, and such has been a prominent feature of Role and Reference Grammar since its onset in the early 1980s. This paper will survey a range of nominalizations across a number of unrelated languages, demonstrating how Role and Reference Grammar, supplemented by the dual head proposal of Bresnan (1997), can provide a empirically rich and theoretically satisfying account of them.