CRLC Seminar Series, ANU, 2009
Normal time: Tuesdays, 11.00am-12.30pm
Seminar Room B, HC Coombs Building (no.9), Australian National University
“Kinship systems of the Port Curtis Coral Coast region of Queensland”
This paper examines and compares the kinship systems of two contiguous language groups in southern Queensland: Gurang and Dappil. Although neighbours, the two groups represent quite dissimilar entities: one, Gurang, large inland, linguistically homogenous, and generally endogamous, and the other Dappil, geographically restricted to a small coastal and riparian environment, and in close contact with similarly disposed language groups of quite different ethnic composition. From an analysis of the structure of their respective kinship systems it is hypothesised that the systems themselves tell us something not only of the history of these language groups but the political imperatives guiding them at the time of sovereignty.
“External pressure prompts language change”
Much research has examined the origins of the Polynesians. This talk examines the developments within Polynesian languages and presents evidence for a dichotomy of kinds of language change based on socio-cultural environment. Not only is change in a language not constant across different modules of language, but it is not constant for different societal circumstances.
“Polynesian paradoxes: subgroups, wave models and the dialect geography of Proto Polynesian”
The Polynesian group is perhaps the best-defined major subgroup in the Austronesian language family. A massive body of innovations, phonological, morphosyntactic and lexical, marks Polynesian off from all other Oceanic languages. The sheer number of innovations points to a very long period of development as a single language after Polynesian separated from its nearest living relatives. It appears that the branch of Central Pacific that became Proto Polynesian (PPn), maintained a high degree of homogeneity for around 1000 years after its speakers first settled the various island groups of eastern Fiji and western Polynesia.
Nevertheless, a number of questions concerning the development and breakup of PPn have not been fully resolved, including the following.
(1) On the evidence of phonological and morphological innovations PPn diverged into two primary branches: Tongic and Nuclear Polynesian. Why is this split not reflected in standard lexicostatistical comparisons?
(2) When did PPn break up? Did the innovations defining the Nuclear Pn subgroup develop late in the Pre PPn period, marking the permanent separation of Tongic and Nuclear Polynesian dialects into independent speech traditions, or did they develop much earlier, persisting as dialectal features during the long period of relatively unified development? If the latter were the case, anyone trying to draw a family tree for Polynesian would face a paradox: the node at which the two major branches of Polynesian separated does not coincide with the end of the Pre PPn period.
(3) How adequate is the family tree (subgrouping) model for showing the historical relationships of the Polynesian languages? Dyen (1981) and Rensch (1987) have proposed that we abandon the family tree model for this purpose and view the Polynesian group, and particularly the speech traditions of western Polynesia, as a dialect complex within which it is difficult or impossible to define discrete subgroups. Does this proposal have merit?
To tackle these questions one needs to reconstruct the dialect geography of the early stages of Polynesian, carefully examining the distribution of innovations and borrowings that can be associated with this period, and to see how far the linguistic evidence correlates with the early history of Polynesia as indicated by archaeological evidence. This paper seeks to proceed along these lines, building on previous work.
“The Austroasiatic Central Riverine Hypothesis”
Recent attention has focused on the question of localizing the Austroasiatic homeland, especially interdisciplinary inquiries seeking correlations between linguistics, genetics, and archaeology. Among the various suggestions that have been offered over the years, one can distinguish three essential ideas:
- a western origin, in northern India or in the vicinity of the Bay of Bengal
- a northern origin, in central or southern China
- a central origin, within Southeast Asia
The first of these was advanced on the basis that Munda languages are morphologically archaic; the second view has found support in supposed Austroasiatic loans into Old Chinese; while the third is mostly a centre-of-gravity argument that finds no bases for locating the ancestors of the Mons, Khmers and others outside of their known historical locations. In recent conference presentations Gerard Diffloth has argued that the Austroasiatic lexicon rules out a temperate zone (i.e. China) origin in favour of a tropical homeland, which would narrow the possibilities to 1. or 3. above. But how can one build further on this insight? I argue that the comparative analysis of lexicon and phonology suggests a flat or rake-like family tree, consistent with an ancient dialect chain strung out along a distinct geographical corridor. The mainland of Southeast Asia offers just such corridors in the form of north-south running river valleys, especially the Mekong, Chao Phraya and the Irrawaddy. A parsimonious analysis of possible migration routes out of such a homeland suggests a zone centred on the middle Mekong, in the lowlands of northeast Thailand and southern Laos.
Such a model anticipates that the ancient Austroasiatics were basically a riverine society that first spread along waterways, adapting secondarily to upland environments, particularly those who moved north and west.
“Proto-Mirndi and the origin of subsections”
Mark Harvey's recent book Proto-Mirndi (2008) is to be celebrated as the first published book entirely devoted to a historical-comparative study of an Australian language family. It is of particular interest since this family is discontinuous, with western languages in the lower Victoria River basin and eastern languages in the Barkly Tableland. The first part of the paper is a general assessment of the reconstructions and methods used.
The second discussion topic is a prominent sub-theme in the book: a reanalysis of the hypotheses and chronologies of the origin and diffusion of subsections (an eightfold sociocentric division) in McConvell (1985 etc), pointing to what Harvey believes is an earlier origin. Other challenges to McConvell's work are also examined, e.g. the hypothesis that diffusion of the terms into Arandic cannot be understood solely in terms of strict 'linguistic stratigraphy' but must take into account the (imperfect) remodeling of loans to conform to expected sound correspondences by speakers. This leads into discussion of the origin and diffusion of sections, a four-fold division found widely in Australia, from which subsections arose by interaction and merger in the upper Daly River area. This is the topic of another talk in second semester.
“Comparative reconstruction of Marrngu”
Karajarri, Mangala, and Nyangumarta are three languages spoken in Western Australia, which have long been assumed to form a subgroup (Marrngu) based on earlier lexicostatistical data. This seminar reviews ongoing research into a historical description of the Marrngu languages. I will be showing data from the languages themselves, and discussing issues that do not support a scenario in which these three languages stemmed from a single ancestor. These include the lack of clear phonological innovations, and the prevalence of shared phonological and morphological retentions from a higher-level subgrouping.
“Researching Contact in the History of English”
While the creolization hypothesis has largely been abandoned, debate continues about what role contact played in the history of early English. This is partly due to ongoing disagreement about which theoretical approach can best explain language contact. A recent hypothesis claims that Scandinavian contact caused English to be substantially simplified. McWhorter (2002) draws on numerous examples in order to argue that English has an unusual loss of overspecification compared to other Germanic languages. A regional analysis of the loss of three of these features was completed using the Penn-Helsinki (PPCME2) corpus. This showed no relationship in early Middle English between contact and the losses proposed as examples of simplification. Van Coetsem's framework is used to explain why these losses were not contact-induced while other important changes in the same period were. This study shows the importance of direct textual evidence from corpus analysis and the benefits of Van Coetsem's speaker mechanisms framework compared to Thomason and Kaufman's system-oriented framework. It also shows how substantial contact-induced change in early English can have coexisted with the continuous transmission of the language.
“Pre-Indo-European Europe and the spread of Indo-European: state of the art and current discussions”
In the typological literature Europe ranks low on the scale of linguistic diversity, whereas, for instance, Aboriginal Australia is usually counted among the linguistic hotspots of the world. However, if the obvious explanation for Europe's paucity in typological diversity lies in the dominance of Indo-European languages, the question arises what Europe looked like before the spread of this language family. This is relevant for at least two reasons.
First, linguists have wondered about the role language contact played in the constitution of the Indo-European daughter languages, and whether this may explain the significant differences between Indo-European languages from different branches. Second, it could shed some light on the origin of some of the lost and unaffiliated languages of Europe, such as Etruscan, Iberian and Basque.
The first part of this talk will explore different perspectives on the linguistic map of pre-Indo-European Europe. I will argue for a scenario that assumes that after the last ice age Europe was initially not high in linguistic diversity, but that a relatively diverse patchwork of languages probably covered the continent at the dawn of the Indo-European expansion, several millennia later, which was subsequently superseded and partially soaked up by the Indo-European languages.
In the second part the talk will discuss the question of how Indo-European spread across the European continent, arguing mainly against a view of demic diffusion, and for a scenario based on language shift, for which the spread of the Pama-Nyungan languages in Australia offers striking parallels.
“What 19th-century Leicestershire dialect and Polynesian L2 English have in common: Multiple sources for some linguistic features on Palmerston Island”
Palmerston Island is a tiny island in the Cook Islands group, approximately 600m across, and with a population of 54. It was settled in the mid-1800s by William Marsters, of Leicestershire, England, and his three Cook Island wives. There may also have been a Portuguese man on the island for the first 10 years or so. The inhabitants today are the descendants of William Marsters and consider themselves to be English. They speak a dialect of English: most of them speak no other language.
The island has been very isolated - there is no regular transport to or from it, and it is 400 km away from the closest other inhabited island.
Nowadays around 30 yachts visit in total each year for a few days each, and detailed records of these are kept. This means that it is possible to (a) record every local speaker of Palmerston Island English, and (b) track all external influences on the language, making Palmerston Island a wonderful opportunity for studying language variation and change in small mixed-origin communities.
In this seminar I will describe some of the features of Palmerston Island English and discuss the possible sources of these. For some features I will show that they could equally have come from Leicestershire English or from Polynesian English, and I will suggest that there is no need to treat them as having a single source. I will also outline some preliminary findings about change and stability of the dialect over time, based on a comparison of my data, collected this year, with data from 20 years ago, and written texts (letters, etc) from the 1950s.
“Western Oceanic revisited: interpreting shared innovations”
The Western Oceanic languages are a collection of Oceanic Austronesian languages located in Papua New Guinea and the northwestern Solomons.
Nevertheless, a number of questions concerning the development and breakup of PPn have not been fully resolved, including the following.
In my PhD thesis, published in 1988, I argued that the Western Oceanic languages shared no common ancestor other than Proto Oceanic. Such unity as they display is due to the fact that they are all descended from an early Oceanic dialect network which remained in place in the Oceanic homeland area (the Bismarck Archipelago) after other Oceanic speakers moved southward and eastward to occupy the previously uninhabited islands of the Pacific. In this talk I will (i) discuss what has emerged from more recent linguistic and archaeological study of the period when (we infer) Proto Oceanic and early Western Oceanic were spoken; and (ii) examine the ways in which the evaluation of shared innovations as markers of shared ancestry has changed in the last 20 years, and re-evaluate some of the innovations that I wrote about in 1988.
An activity of the