CRLC Seminar Series, ANU, 2010
Normal time: Wednesdays, 4.00pm-5.30pm
Baldessin Precinct Building (no.110), Australian National University
“Philology and the reconstitution from archival sources of NSW south coast languages”
The languages of the New South Wales South Coast consist of four, possibly five, languages: Dharawal, Dharumba, Dhurga, Djirringanj and Dhaawa. These languages are part of what Schmidt (1911) classified as the Yuin-Kuri (yuwinj and kuri �man� in the southern and northern languages respectively) language group. Because my analysis is based on exclusively archival written material (published and unpublished) that was collected between the 1830s and 1900, my language work is based on an ongoing philological analysis of the historic material. In this seminar, I would like to proffer some of the challenges and victories that present themselves when working from and with archival material. Identifying quality of rhotic sounds in historic written material is one of the challenges; detecting morphological patterns, and functions, in blocks of texts is just one of the joys of working with this material.
The South Coast languages (SCLs) are closer in grammatical and lexical form to some of the other coastal languages of NSW like Darkinyung and Gadhang, rather than the central NSW languages Wiradjuri, Ngiyampaa and Gamilaraay. Whereas Dharawal has pronominal enclitics that can be found on the first word in a clause, there is no evidence to suggest that this is also the case in the other SCLs. Phonologically, all South Coast languages have word-final stops, laminal contrast in word-initial position, and I propose that there is evidence for a series of retroflex phonemes.
“Loanword stratigraphy, lenition and the diffusion of subsections in the Kimberley”
Results from McConvell (2009) and more recent work in the project Dynamics of Hunter-Gatherer language show levels of loanwords of over 40% in Eastern Ngumpin languages like Gurindji, which is significantly greater than in Western Ngumpin, Nyulnyulan and North and South American languages we are working with. Eastern Ngumpin has high numbers of loans from Non-Pama-Nyungan languages to the north, Western Mirndi and Jarragan, which range from around 20% to almost zero. There is also fairly significant borrowing internally within Eastern Ngumpin from Western Ngumpin, for which the evidence includes absence of medial lenition (*p,*k>w; *ty>y; *rt>r), which is analysed as a regular sound-change in Eastern Ngumpin Medial lenition is an important areal phenomenon covering southern Jarragan languages as well as Eastern Ngumpin. other language. Apprent lenition in Bunuban however is rather a property of loanwords into Bunuban languages, some from Western Ngumpin which itself does not have lenition. This points to the existence of a mystery leniting language at one time in the Central Kimberley.
Harvey (2008) on the Mirndi language family, reviewed in a previous CRLC seminar also has a large section on subsection diffusion, mainly supporting the hypothesis of McConvell (1985) but also introducing revisions such as tracing subsection roots to proto-Mirndi and interpreting lenition as a sporadic process due to word frequency. Because of his geographical focus, Harvey did not take much account of Kimberley subsection terms. Here the analysis of lenition and stratigraphy in the Kimberley in the first part of the paper is brought to bear on this issue, painting a different picture of the diffusion of subsections.
“Proto Palaungic phonology: the problem of reconstructing vowel lengths and qualities in a system undergoing restructuring”
The Palaungic languages, spoken mainly in the Shan state of Myanmar and border areas of Laos and China, represent special problems for Austroasiatic reconstruction. Most have lost the Austroasiatic short-long vowel contrast and developed tones and/or voice registers, with strong parallels to the main contact languages Burmese and Shan. 4 or 5 sub-groups are apparent, and reconstructions have been published for 2: Waic and Palaung-Riang, and I have prepared a Proto Palaungic reconstruction using data from all sub-groups. The proto-consonant system is straightforward, but the vocalism is challenging. It is difficult to systematise the vowel correspondences, and it is apparently necessary to reconstruct some peculiar asymmetries in the distribution of length contrasts, indicating a system in transition. And although diphthongs are common (even prolific) among the daughter languages, it is not clear whether diphthongs should be reconstructed at all for Proto Palaungic, or how to model the classification to reconcile such a phonological history. A solution is offered which depends to some extent on external comparisons.
“Types of language contact with Old Norse”
English historical studies of language contact have so far taken little note of general contact linguistics. Some of those that have, suffer from a lack of language-specific textual and historical knowledge (e.g. Thomason & Kaufman 1988: ch. 9 on Old Norse and Old French influence). Other studies, based on textual and statistical material and on the historical facts, have led to a more differentiated understanding of foreign influence than provided by established handbooks (see Lutz 2002 for extreme superstratal influence of French).
This paper on Old Norse influence in early England intends to show (1) that the Old English lexical evidence (Peters 1981, Townend 2002) neatly reflects the situation in late Anglo-Saxon England, namely that of Danish dominance much beyond the Danelaw (Rumble 1994). The evidence (e.g. E. earl, law) can be characterized as mainly superstratal, similar to Old French influence on post-Conquest English (E. duke, justice). It is mostly from West Saxon texts and thus from outside the Danelaw.
Moreover, this paper will discuss (2) Old and Middle English lexical and structural evidence for various degrees of Old Norse adstratal and substratal admixture for different dialects. Due to the dialectal imbalance of the Old English evidence, much of this influence turns up only in early Middle English texts. Also, due to the close genetic relationship between Old English and Old Norse, this influence is often less obvious because it may be interpreted as either Old Norse or Old English material (e.g. the pronominal form she, first attested in the former Danelaw in 1140). Or it may be reflected as a more speedy morphological and syntactic development in the most heavily Scandinavianized Northern dialects (e.g. the present-tense marker -s and a tendency towards generalized VO-order in the Lindisfarne Gospels; cf. Lutz 1991: ch. 2, Kroch & Taylor 1997).
The characterisation of such examples for likely Norse influence as evidence for creolisation is problematic (Poussa 1982, Hansen 1984). But on the basis of a cautious application of terms from general contact linguistics, we are likely to arrive at a more differentiated understanding of Old Norse influence on English in different dialects of the early Middle Ages.
“Vasconic features in West Indo-European: Structure and lexicon”
In the structure part of the lecture three topics will be treated, of which two have been dealt with before, one is new: (1) vigesimality in Danish, Insular Celtic, and Western Romance; (2) two-copula syntax in Old English, Celtic, and Western Romance; and (3) first-syllable accent in Germanic, Celtic, and Italic.
In the lexicon part of the lecture, too, three topics will be treated, two of them with old and new etymologies, and one new: (1) appellatives, viz. the Wanderwoerter iron and silver, and direct borrowings such as Irish ainder 'young woman' (MFrench andre 'woman', Ital. landra 'prostitute' etc.), Lat. caseus 'cheese', and E bush, Rom. bosco/bois; (2) toponyms such as Isar/Is�re, Arn-/Earn- names, Cannes/Canossa, and esp. Lech/Loire; (3) under the heading "Lexicon and culture" the sporadic use of expressions meaning literally 'to-night' instead of more common expressions meaning 'to-day' in Germanic and Celtic.
All of these Western structural and lexical features have no parallels in the Eastern Indo-European languages but do have parallels in Basque. Therefore, a case can be made that they are original in very early Basque and were carried into Western Indo-European from a prehistoric Vasconic substratum. In the conclusion reference will be made to genetic studies supporting the Vasconic Theory, namely the idea that Western Europe was repopulated after the last glaciation by a northward expansion from the Basque Urheimat near the Pyrenees.
“Variation and agreement in Afro-Bolivian Spanish DPs”
In this presentation we analyze plural marking in the Afro-Bolivian Spanish (ABS) Determiner Phrase (DP), in order to establish the process underlying its variation.
Plural marking variation is examined following recent approaches to syntactic variation (Adger and Smith 2005, Adger and Trousdale 2007) whereby two different syntactic inputs will nevertheless yield semantically equivalent outputs.
Several works have focused on similar phenomena in other Spanish/Portuguese dialects from a sociolinguistic perspective (Cedergren 1973, Terrell 1976, Poplack 1979, 1980, Braga 1977, Guy 1981, Sherre 2001). While in some of these studies phonological and stress factors seemed to play a major role, this paper illustrates that grammatical features are the actual locus of variation in ABS.
Building on current syntactic assumptions (Chomsky 2000, 2001; Bernstein 2001; Longobardi 2001; B�jar 2003; Pesetsky and Torrego 2004; Adger et al. 2008), we argue that the presence of valued number features in ABS is underspecified/reduced in comparison to Standard Spanish. Uninterpretable phi-features in agreement relations are redundant and their absence is grammatically costless. Such hypothesis is congruent with previous studies on impoverished agreement in ABS DP (Lipski 2007).
This study looks into ABS number marking moving from a variationist approach into current formal syntactic theory, in line with recent lines of research linking sociolinguistic methodology and syntax.
“Variation in the proto-typical sentential negative pattern across Quechua dialects”
This is a presentation of my work-in progress on the analysis of data about negative sentences collected during my fieldwork in Quechua speaking locations in South America. In the literature, Quechua is commonly treated as essentially being uniform in its method of negation, but I have found a good deal of variation. Several works have described Quechua sentential negation for a particular dialect but guided for the general assumption of the homogeneity downplaying differences. Others have focused on the meaning of the markers involved in cross dialectal Quechua negation but highlighting discrepancies in the author�s descriptions rather than analyzing the data themselves. The motivation to carry out my fieldwork was the conviction that a study of the synchronic situation of the phenomenon of negation is essential to understand better its evolution. My research aims to explain the development of syntactic negation in Quechua.
The sentential negative pattern in this group of languages is far from being uniform. In fact, Quechua exhibits quite diverse variation, which my research endeavors to describe systematically in order to account for the changes operating in its negative system. Only once this description is made, we can hope to understand how the different patterns have developed historically and what changes are underway currently. In this seminar I will present how the proto-typical Quechua sentential negative pattern �which I label �embracing negation�� works vis-�-vis existing variation in other groups of dialects. Embracing negation is the one that is usually presented as the pattern for Quechua generally. The classification I will propose aims to systematize the encountered variation in order to understand how Quechua dialects have split into two broad alternative negation patterns. In doing so, we can observe ongoing changes through attending to synchronic variation. I will discuss the possible role of language contact. My findings also point to highlight a revival of Quechua in places where it was previously considered endangered.
“Interpreting the results of typological analysis: geography and phylogeny”
While family trees are built on the basis of the comparative method, a methodology designed to maximise the signal from inheritance, it is also possible to perform typological analysis on language structures and then cluster the results. This talks examines the results of different kinds of typological cluster analysis applied to languages in different social contexts, and argues that we should consider inheritance, as determined by the comparative method, as only a minor contributor to the clusters arising from typological analysis.
“It takes two to tango and three to huayno: Strange lo as a (syn)tactic move”
Formal theories have mostly been concerned with either idealised data sets that nicely fit their theory, or an idealised relationship between speakers and their homogeneous speech community. Recent work in Lexical-Functional Grammar has not only addressed language change and/or dialect variation (Vincent 2000), but included microvariation as in nonstandardized variation (Bresnan 1998, Bresnan 2001, Bresnan&Sharma 2007).
My talk is based on my thesis which examines the complex relationship between primary agreement by feature specifying clitics, and secondary object agreement by differential object marking (DOM) in nonstandardized Lime�o Spanish contact data (LSCV) collected in fieldwork. The particular focus is on strange lo, a featureless and invariate form, on extended DOM, on the effects of coocurrence of both, and the relationship to secondary topic and primary object marking.
I will show that syntactic covariation of marked and unmarked forms can be interpreted in terms of semantic and pragmatic strategies particular to contact speakers. Synchronic agreement mismatches arise at the crossroads of multiple contexts based on a diachronically attested struggle of dative and accusative for primary object status in monotransitive clauses.
“Language change in Byzantine Greek: examples from the texts of Anna Comnena�s Alexiad”
Early in the last century historians drew attention to a Byzantine document, namely the Alexiad, written by Anna Comnena (1083-1154 (or 1155 A.D.)) the daughter of the Byzantine emperor Alexius Comnenos (1048-1118) and the woman who rightly acquired the title of the first woman historian. Apart from being the flattering biography of the emperor Alexius Comnenos, the Alexiad is a major source of the historical events of the first Crusade. Anna Comnena�s magnus opus consists of fifteen chapters that narrate the story of Byzantium between the years 1069 and 1118, and most particularly of the period of Alexius� reign (1081-1118). The document under analysis offers an exceptional source for the study of language change in Byzantine Greek. While historians treat Anna Comnena�s accounts on the events of her times with some caution, the language displayed in the Alexiad needs not being questioned: Anna wrote her work in what has been described as pseudo-Classical Greek, a learned language, or, otherwise named in Greek a �p?ast? ???ssa�= ( a made up language), �unknown� to the ordinary people of Constantinople. In the preface of the translation of the Alexiad by E.R.A. Sewter, Anna�s language is described as �an almost entirely mammiform school language�.
This paper aims to analyse Anna Comnena�s Greek and discover, identify and illustrate some examples of language change in Byzantine Greek that have been well disguised behind the �mummiform school language�, used in the Alexiad. Meticulous investigations of the texts reveal areas of change that escaped or could not follow the �made up� Classical form of Greek. This first approach to Anna�s language depicts changes in phonology and in semantics inasmuch as it exposes phenomena of reanalysis the very time they were taking place.
“Diffusion and semantic change in in-law terms in languages of northern Australia”
Affinal (spouse and in-law) terms seem to diffuse widely much more readily than other kinship terms, probably due to the extension of marriage networks and marriage practices. This paper looks at some cases of such diffusion in Northern Australia. The primary case concerns the root ramparr which starts from the non-Pama-Nyungan Worrorran languages in the North Kimberley, spreads to Nyulnyulan, then changes to lamparr due to a regular sound change of lateralisation in early Ngumpin-Yapa (a Pama-Nyungan sub-group to the south). As it enters Mudburra in the eastern Victoria River District it has an -a augment added as is regular in Mudburra with former consonant final roots. In this new form lamparra it spreads rapidly across the savannah belt of the Northern Territory in the Northern Territory along communication lines established by the cattle industry. Along with these sound changes there are also meaning changes to the root. The original meaning seems to have been �screen� or �barrier� in Worrorran, extended to mean avoidance relations in general, then narrowed in some languages to mean wife�s mother, the prime avoided affine. In Nyulnyulan it retains the WM meaning initially but in some languages the meaning shifts to wife�s father. The WF meaning (self-reciprocally also son-in-law) is the one adopted by other languages and diffused through the NT. The talk attempts a chronological stratigraphy of the changes in the root and their diffusion; and an explanation of the change of meaning from mother-in-law to father-in-law.
Several other roots and their patterns of diffusion and inheritance are also briefly considered, together with possible correlations with marriage systems. First, the spouse sibling/in law term ngumparna (and variants) in the Kimberley and across the Northern Territory, and its relationship to a much more widespread Pama-Nyungan spouse term nyupa (and variants). Finally another diffusion of a term tyamV(ny) �mother�s father/mother�s brother�s son/spouse� across the non-Pama-Nyungan languages of the Kimberley and across the savannah belt of the Northern Territory.
“Tone Sandhi Microvariation in Lishui Wu and Adapting the Comparative Method to Tone Sandhi”
ABSTRACT The phonological changes that occur on the tones of two-syllable words in Lishui Wu are very complex, even for a Wu variety. The tone sandhi typology and the tone shapes (both isolation and sandhi tones) of the three speakers analysed in my doctoral thesis do not vary much, but the realisations do. I will present firstly the three different tone sandhi patterns of the speakers, and then present a preliminary analysis of how they may be explained from a diachronic perspective.
The three speakers come from three different generations, and their sandhi patterns probably represent a sort of tone sandhi change, rather than geographically-based variation. A detailed analysis is difficult, because the three patterns do not represent a simple A > B > C change. The older speaker cannot necessarily be said to speak a direct antecedent of the other speakers� varieties. Further difficulties arise from the lack of a tested methodology to apply the comparative method to reconstruct polysyllabic tone sandhi. I will propose a more reliable, yet more time-consuming method of applying it.
“Akkadian as a Lingua Franca: Language Use and Linguistic Identity in the Ancient Near East”
The 2nd millennium B.C. is a period where a marked increase in organised trade relations and international contact over a wide area can be observed in the archaeological and philological record of the Ancient Near East. The language used to facilitate communication in this internationalised environment is an Akkadian dialect, in use from Cyprus to Anatolia, Egypt and Syria. The project focuses on the questions of how, when and why Babylonian became the main lingua franca of the Ancient Near East and how this process affected the local language communities and their linguistic identity and self-perception. The data comprises documents written in cuneiform script unearthed in archives throughout the Near East, mainly in Egypt, Anatolia and along the Levantine coast. Two main foci will be relevant for the analysis of the data: on the one hand the archaeological material and the context in which the texts were written, and on the other hand the linguistic features of the texts in question, especially with regard to bilingualism, dialectal forms and source language interference.
An activity of the