Tuesday Seminar Series

The Linguistics Department Tuesday Seminar is held in Biomed T208 at the University of Hawai’i at Mānoa from 12:00pm to 1:15PM every Tuesday in both Fall and Spring semesters.

Any topic related to linguistics is welcome. The seminar coordinator is Robert Blust.  If you are interested in giving a talk or would like further information, please contact the seminar graduate assistant, Amber Camp.

***SPECIAL DATE: THURSDAY, JANUARY 26, 2017***

To document a language, one must describe kinship.  Where shall we start? As someone primarily interested in comparing systems within languages, I express an explicit interest in affines, related by marriage(s) and the lexical system that connects these folk to consanguines, related by birth(s). There are four basic types of kinship systems based on terms used to cover the parents and their siblings: Generational (aka “Hawaiian”), Lineal (aka “Eskimo,” US), Bifurcate merging (aka “Iroquois,” popular in Melanesia), Bifurcate collateral (aka “Senegalese,” Latin). Jabêm poses an interesting challenge to this typology:  sa ‘aunt/uncle.’

This study explores affinal typology that emerges from comparing Kâte & Jabêm:

  • non-Oceanic languages of South Bougainville nearly lack explicit affine terminology. Hage (2004) classes them as Dravidian (a type of Iroqois).
  • Oceanic languages of Bougainville have richer affine terminology including Petats : súhan  ‘parents of child’s spouse.’ (Blackwood 1935:590).
  • Kâte the Trans New Guinea language nearest Finchhafen, PNG has two bi-affinal (involving  two marriages) terms:  borâ ‘husband’s brother’s wife’; gorec ‘wife’s sister’s husband.’ (Dempwolff 1925).

In current constraint-based models of phonological learning, learning the legal (and illegal) sound sequences in one’s language (i.e. phonotactics) facilitates learning alternations. However, there is still little empirical evidence for this. I first show that alternation learning in an artificial grammar experiment is facilitated when the phonotactics in the lexicon match the alternation. This supports current models of phonological learning. What does this mean for languages with a mismatch between phonotactics and alternations? In the second part of this talk, I present corpus analyses as well as computational learning simulations of two such languages – Korean and Turkish. I show that in both languages the reported mismatches between phonotactics and alternations are superficial. This undermines assumptions in previous analyses of these patterns. Instead, I argue that there is a bias to maintain similar generalizations in phonotactics as well as alternations. This further supports the claim that learning phonotactics aids in learning alternations, ultimately arguing in favor of constraint-based models of phonology that account for these two generalizations using a single mechanism.

***SPECIAL DATE: THURSDAY, FEBRUARY 9TH, 2017***

The phonological lexicon, the list of phonological forms for a given language, is of central importance to both phonological and psycholinguistic theory. In this talk, we present a novel approach to the study of the phonological lexicon via complex network modeling. This modeling relies on the notion that each word in the lexicon is connected to its nearest phonological neighbors. By connecting all words in this manner, the lexicon can be examined as a network of interconnected nodes, and the tools of complexity science can be applied. By comparing the lexicons of nineteen genetically and typologically dissimilar languages to over four thousand pseudo-lexicons randomly generated under controlled conditions, it is possible to determine the provenance of various features of lexical structure. We confirm that, in accordance with linguists’ intuitions, phonotactics play an important role in determining the structure of lexicons. However, above and beyond the effects of phonotactics, we find evidence for two higher-level cognitive constraints operating over the lexicon: a pressure to form new words out of subcomponents of existing words; and a pressure against the existence of phonologically similar words. These pressures are hypothesized to be of functional importance for language acquisition, production, and comprehension. Future work and extensions to other domains are discussed.

Comparative dictionaries are rare, as they require knowledge of the phonological histories of numerous related languages, together with years of intensive work in assembling cognate sets and providing reconstructions that show how these sets originated from unitary forms.  The Austronesian Comparative Dictionary (ACD) began as a 59-page article published in 1970.  Work leading up to it continued intermittently through the 1970s and 1980s, and culminated in an NSF grant from 1990-1995 which combined all previously published material with new material in a unified online comparative dictionary with about 2,500 base entries and many more affixed forms.  There was then a 15-year lapse when little further work was done on it.  Starting in 2010 a chance meeting of Robert Blust with Stephen Trussel reinvigorated the work, which has expanded dramatically since then both in scope and in computational sophistication.  A report on this exceptionally productive collaboration was given as a Tuesday seminar talk on March 6, 2012, and published in the December, 2013 issue of Oceanic Linguistics.  This talk aims to show how much further growth has taken place in the five years since the first talk, and in doing so to give some idea of the scope, complexity, computational sophistication and scholarly value of a work that is now going on 47 years in the making, and ranks as the largest research project ever undertaken in the study of the Austronesian languages.

NOTE: Please bring your laptops, as part of this talk will require you to use them to call up data samples from the ACD (www.trussel2.com/ACD).

In this talk, I use traditional constituency diagnostics such as coordination and adverb placement to investigate the status of the verb phase in Austronesian. Conservative Malayo-Polynesian languages are shown to treat the verb and Agent of a Non-Actor Voice clause as a constituent which excludes the Patient (in agreement with earlier analyses of languages such as Tagalog and Malagasy). This stands in strong contrast to (both accusative and ergative) languages of mainland SE Asia, most of which show relatively good evidence for a traditional Verb Phrase that includes the verb and Patient while excluding the Agent. 

Expanding on earlier work, I argue that the unusual grouping of transitive Agent with the verb is accounted for in Austronesian by the historical reanalysis of possessors as agents. Possessors naturally form a constituent with the noun phrases they modify and thus the Possessor/Agent forms a constituent with the verb after the N>V reanalysis. This leads us to an interesting prediction regarding Puyuma and Tsou under Ross’s (2009) Nuclear Austronesian hypothesis. If only the realis paradigm consisting of Patient Voice *-en, Locative Voice *-an, and Circumstantial Voice *Si-  derives historically from the reanalysis of nominalizations, we do not expect the unusual constituency of Non-Actor Voice clauses in Malayo-Polynesian languages to show up in the irrealis paradigm, which never passed through this nominal stage. 

Evidence from Puyuma and Tsou points towards the existence of just such an asymmetry. Teng (2008) strongly suggests an unmarked word order in Puyuma Non-Actor Voice clauses that is highly unusual for Formosan and Philippine languages: the pivot (or NOM argument) regularly precedes the transitive agent. The unmarked status of this order was supported by experimental elicitation in recent fieldwork. Crucially, however, the unmarked Pred GenP NomP order of NAV clauses in Nuclear Austronesian languages also emerged in Puyuma in clauses with nominal predicates. This can be interpreted to support the Nuclear An Hypothesis as it represents an additional syntactic parallelism between nominal predicates in extra-Nuclear An languages and the canonical declaratives of Nuclear An languages.  Beyond these particular arguments specific to Austronesian, I aim to address the wider relevance of word order and clausal constituency to solve historical questions.

Abstract:
Bati (btc – ISO 639-3) is spoken in Cameroon by a relatively small community of approximately 500 to 1000 speakers. The Bati speech community is originally located in three distinct villages namely Kelleng, Mbougue and Nyambat, all belonging to the same Canton within the Sanaga Maritime Division, Littoral Region of Cameroon. Each of these villages is host to distinct sets of language repertoires. This variation in language use, though not a critical impediment to mutual intelligibility across speakers of the overall Bati community, is a leverage to micro-identity construction and sub-community demarcation, as much as it nurtures sociolinguistic meta-discourse relating to language purism and correctness. To this effect, each village overtly claims to speak a separate tongue which is named after each respective village (Kelleng, Mbougue and Nyambat).

Given a language repertoire in usage in the Bati speech area namely in the Kelleng village, this study will seek to investigate the likeliness of a given language variety to be resorted to by a speaker in the course of a naturally occurring conversation, as well as its frequency. The study is based on a sociolinguistic experiment conducted over five-hour orally recorded speech documenting two sessions of naturally occurring conversation in Kelleng, one of the three villages of the Bati Canton. All participants featuring in both sessions are female speakers living in Kelleng but originating from Kelleng as well as from various neighboring communities. Participants are regularly involved in the same social networks, which made it possible to follow them from one social activity to another, namely farm cultivation and cassava paste preparation. Both sessions were video-recorded and eventually annotated for both the language repertoire and the socio-cognitive network in which each bit of turn taking (as reflected by the nearly arbitrary segmentation of annotations using ELAN) may be inscribed (public, private, gossip and business). Data frames have been further created for each recording. R is used as a tool to compute both probability and frequency of language variety using linear regression analysis. Based on the scope of the experiment, results show predictable likelihood for choice of specific language varieties depending on the conversation momentum.

Introduction & Methods:
1.    Introduction
Bati (btc – ISO 639-3) is spoken in Cameroon by a relatively small community of approximately 500 to 1000 speakers (Ngué Um et al., on-going language documentation project ). The Bati speech community is originally located in three distinct villages namely Kelleng, Mbougue and Nyambat, all belonging to the same Canton within the Sanaga Maritime Division, Littoral Region of Cameroon. Each of these villages is host to distinct sets of language repertoires. This variation in language use, though not a critical impediment to mutual intelligibility across speakers of the overall Bati community, is a leverage to micro-identity construction and sub-community demarcation, as much as it nurtures sociolinguistic meta-discourse relating to language purism and correctness. To this effect, each village overtly claims to speak a separate tongue which is named after each respective village (Kelleng, Mbougue and Nyambat).

In addition to being a linguistically diversified community, Bati is an inherently multilingual group. This state of affairs is due to geographical, social and historical factors. Geographically, Bati territory is situated at the junction of four major yet dominant speech groups namely Basaa (bas – ISO 639-3), Bisoo/Bakoko (bkh), Eton (eto – ISO 639-3) and various sub-communities belonging to the Yambasa group such as Nubaca (baf – ISO 639-3) and Mbule (mlb – ISO 639-3). Geographical proximity with neighboring groups yields regular social interaction between Bati and these communities; a typical illustration of this interaction being inter-ethnic mariages. At the historical level, German colonisation has instituted the use of Basaa as a medium of instruction in schools, and as a means of diffusion of the Holy Gospel.

These factors have brought a repertoire of various speech varieties into the daily language usage of the Bati people, to the extent that every adult speaker who has lived continuously in either of the villages which make up their speech area is fluent in a least five different language varieties.

Bati speakers enjoy a unique “gift” for multilingualism among their broad geographical location which overlaps between the Mbam-Nkam, the Basaa and Beti speech areas. However fluency in more than one language variety is a common skill which many individuals beyond the Bati community have naturally acquired or developed over the course of their social life elsewhere in Cameroon and in Africa. Indeed in many regards, it is safe to say that “multilingualism has been a fact of social life in Africa for a very long time” (Whiteley 1971: 1).

Being a hub of extensive social multilingualism, Africa has inspired a wealth of scholarly inquiries within the field of sociolinguistics. As Di Carlo  (2015) has noted it, early research works in this connection have been mainly concerned with urban settings (Juillard 1995; McLaughlin 2001; Myers-Scotton 1993). In the recent years however, there has been a growing interest for investigating rural multilingualism, as illustrated by Di Carlo (2015), Di Carlo & Good (2014), Connell (2009), Lüpke (2010a, 2010b), and Cobbinah (2010). Quite incidentally, both early and recent scholarly undertakings focussing on African multilingualism have relied on mainstream sociolinguistic and ethnographic approaches which fall within Irvin & Gal (2000)’s “iconization process”. Through this process, linguistic features and the social phenomena which index them appear to be in an iconic relationship. Iconization processes further lead to a methodological constraining of the inquiry into a bi-dimensional bundling: multilingualism being always stated as a default dependent variable for which such predictors as social structures, ideologies, identities or similar ethnographic variables, are called into play.
Without downplaying the importance of social and ethnographic variables in nurturing and shaping social multilingualism in Africa, inasmuch as multilingualism is “a fact of social life” (Whiteley 1971: 1), it is no less relevant to appraise its dynamics by tracking down the on-line distribution and choice of varieties through the randomness of the momentum of naturally occurring conversation. The issue of “who speaks what language to whom and when” (Fishman 1965), though not entirely overriden in the present study, is reconsidered in such a way that the outcome of our sociolinguistic experiment is stated in random rather than in constant terms.

2.    Research Question, Methodology and results
Given a language repertoire in usage in the Bati speech area namely in the Kelleng village, this study will seek to investigate the likeliness of a given language variety to be resorted to by a speaker in the course of a naturally occurring conversation, as well as its frequency.
The study is based on a sociolinguistic experiment conducted over five-hour orally recorded speech documenting two sessions of naturally occurring conversation in Kelleng, one of the three villages of the Bati Canton. All participants featuring in both sessions are female speakers living in Kelleng but originating from Kelleng as well as from various neighboring communities. Participants are regularly involved in the same social networks, which made it possible to follow them from one social activity to another, namely farm cultivation and cassava paste preparation. Both sessions were video-recorded and eventually annotated for both the language repertoire and the socio-cognitive network in which each bit of turn taking (as reflected by the nearly arbitrary segmentation of annotations using ELAN) may be inscribed (public, private, gossip and business). Data frames have been further created for each recording. R is used as a tool to compute both probability and frequency of language variety using linear regression analysis.
Based on the scope of the experiment, results show predictable likelihood for choice of specific language varieties depending on the conversation momentum.

3.    References
[1] Cobbinah, Alexander, The Casamance as an area of intense language contact: The case of Baï- nounk Gubaher. Journal of Language Contact», THEMA III, 2010; 175–201.
[2] Connell, Bruce, Language diversity and language choice: a view from a Cameroon market. An-thropological Linguistics; 51, 2. 2009; 130–150.
[3] Di Carlo, Pierpaolo, Multilingualism, solidarity, and magic: new perspectives on language ideologies in the Cameroonian Grassfields. In: Atti del XLVI Congresso Internazionale della Società di Linguistica Italiana – Siena 27-29 September 2012: 290 – 304.
[4] Di Carlo, Pierpaolo & Jeff Good. What are we trying to preserve? Diversity, change, and ideology at the edge of the Cameroonian Grassfields. In Peter Austin and Julia Sallabank (eds.) Beliefs and Ide- ologies. Oxford:Oxford University Press; 2014; 231–264.
Fishman, Johsua A. Who Speaks What Language to Whom and When?. Linguistique, Vol. I, No. 2; 1965; 67–88.
Irvine, Judith T. & Susan Gal (2000). Language ideology and linguistic differentiation. In Kroskrity (2000b); 35–84.
Juillard, Caroline. La vie des langues à Ziguinchor, Sociolinguistique urbaine. Paris, Presses du CNRS; 1965.
Lüpke, Friederike. Language and identity in flux: in search of Baïnounk. Journal of Language Contact THEMA 3; 2010a; 155–174.
Lüpke, Friederike. Multilingualism and language contact in West Africa: towards a holistic perspective. Journal of Language Contact» THEMA 3; 2010b; 1–11.
McLaughlin, Fiona. Dakar Wolof and the configuration of an urban identity. Journal of African Cultural Studies 14, 2;  2001; 153–172.
Myers-Scotton, Carol. Social motivations for code-switching. Evidence from Africa. Oxford,: Clarendon Press; 1993.
Whiteley, Wilfred H. Introduction. In W. H. Whiteley (ed.), Language Use and Social Change. Problems of multilingualism with special reference to Eastern Africa. Oxford: Oxford University Press; 1971; 1– 23.

**A special student showcase featuring four of our current Ph.D. candidates!**

The perception of words stereotypically associated with younger and older Korean speakers

Jonny Kim (Linguistics, UH Mānoa)

Three findings demonstrate how probabilistic information in language is organized in the cognitive system together with social indices, and how this storage social information affects lexical access. First, words are better recognized when spoken by a talker whose age matches the age associated with the word. Word-stereotypes wield an additional influence on this effect above and beyond that of distribution. Second, recognizing age-associated target words is also influenced by age-neutral talkers; use of an age-related sociolinguistic marker embedded in a preceding utterance, indicating that activation of social exemplars spreads over to guide the processing of a subsequent word. Last (in progress), the effect of congruence between talker age and ‘word age’ is also observed in online processing, as evidenced by eye-tracking data.Therefore, associations are encoded at a representational level, and its effect occurs during lexical access, not as a consequence of post-access integration between acoustic signals and contextual information.


 

Topological relations in Pohnpeian

BRADLEY RENTZ (Linguistics, UH Mānoa)

In this talk I present a more nuanced understanding of topological relations in the Pohnpeian language. The data were collected via the BowPed toolkit with five Pohnpeian L1 speakers. The data were then categorized with the help of evolutionary classification tree modeling. The results show that the two prepositions in Pohnpeian, nan and ni should be redefined in terms of topological relations as ‘containment’ and ‘attachment’ respectively. Likewise the meaning of some prepositional nouns are further revised.


 

A featural description of antipassive-type structures

Raina Heaton (Linguistics, UH Mānoa)

In this talk I describe variations on antipassive-type patterns which were observed in a typological survey of 445 languages from all over the world. There were eleven features related to antipassives tracked for the survey, and their interaction describes a wide variety of structures, both antipassive and non-antipassive (at least by most definitions). I begin with a discussion of the features and why they were chosen, and give a basic working definition for antipassive. I then briefly discuss the eight common patterns of features found across the languages in the sample, and provide a series of schematics which illustrate how these features relate to each other. These findings inform our understanding of antipassive-type phenomena, and give us an idea of how frequent different patterns are across a genetically and geographically diverse sample.


 

Tone-intonation interaction in context in Thai

Amber Camp (Linguistics, UH Mānoa)

The acoustic realization of lexical tone is influenced by sentence-level intonation. For example, F0 measurements show that a phonologically falling tone in Thai differs in sentence-medial and sentence-final contexts. Sentence finally, when it is overlaid with a falling intonational contour, F0 falls further and begins its decline at an earlier timepoint than when in sentence-medial position. This categorical perception study, which includes identification and discrimination tasks, uses two nine-step continua of target words created with naturally produced lexical tone endpoints, presented within two different naturally produced sentence frames. These tasks investigate the perception of high and falling tones in sentence-medial and sentence-final positions in Thai to probe whether intonation on the sentence level influences perception of tones. The results shed light on the interaction of tone and intonation in the perception of natural speech, and offer insight into the mechanism with which language users process suprasegmental information.

Due to Spring Break, there is no talk scheduled for March 28th, 2017.

The next talk scheduled is April 4th, 2017.

As part of my dissertation research, I created a typologically oriented database of 445 languages from 144 language families. This database includes information not only on antipassives and antipassive-type features that particular constructions exhibit in these languages, but also on typological characteristics which are relevant to voice, valency, and argument structure. This talk presents broad typological correlations with antipassives to provide a clearer and more diverse picture of the types of languages which have antipassive structures. I also use the database of antipassive constructions to debunk or confirm specific claims which have been made based on less data with respect to antipassives, alignment, valence orientation, basic word order, syntactic ergativity, and switch-reference.

“A Discourse and Grammar Approach to Interaction in Landscape Ontology Elicitation Sessions in Hawaiʻi” by Catherine Lee

This talk critically analyses the method of oral freelisting as an elitication strategy for research on emic understandings of the natural world. Data derive from an oral freelisting activity directed at local landscape ontologies in Hawaii, conducted in a mix of English, Pidgin (Hawaiian Creole English), and Hawaiian. This elicitation strategy involves asking a sample of members of a community of practice to list as many items belonging to a category as possible. It is worthwhile to critically examine the local linguistic interactions in such data elicitation sessions. As interest in the documentation of Local Ecological Knowledge increases, researchers should reflexively examine their own role in authorization of knowledge. I combine the Discourse and Grammar approach with Interactional Sociolinguistics to answer the following research question: How do the participants and the researcher (me) in this freelisting activity authorize or illegitimize their own knowledge and their interlocutor’s knowledge? Findings include the use of code-switching, hedging strategies, epistemic verbs, and laughter when they deem authentication strategies to be necessary. The discourse structure of the data, which would not be available in a written freelist, indicates that certain items require authentication in interaction, while others are taken for granted. Although time-consuming to analyze, these oral freelists provide useful data to judge the epistemic force and evidential nature of various shared cultural knowledge.

“The status of language nests worldwide” by Eve Okura

There are approximately 300-350 language nests in the world. These daycares are established as methods to revitalize endangered languages. In this study, 15 interviews were conducted with language nest works, administrators, teachers, and linguists involved in language revitalization. Seven language nests were visited in person. These interviews and visits addressed questions including but not limited to: 
  • What does it take to establish a language nest?
  • What does it take to maintain one?
  • What are some of the differences between language nests that continue and those that collapse within a year or two?   
  • What resources do language nests have?
  • How are they funded?