When? Wednesdays 2:00-3:30PM

Where? 1303 Dwinelle Hall

What? We are a working group dedicated to the critical examination of methodologies in language documentation, description and revitalization, as well as to the linguistic and ethnohistorical analysis that falls out from that work. Our aim is to learn from and ultimately improve upon methods for carrying out more rigorous, insightful and ethical linguistic and cultural fieldwork, and to help researchers implement those methods.

How? Fieldwork Forum is made possible through a Working Group Grant provided by the Townsend Center for the Humanities at the University of California, Berkeley.

Who? FForum is organized by Julia Nee and Zachary O'Hagan. We welcome all those interested in linguistic fieldwork, with all levels of experience, including those in other departments.

See a list of our past talks here

Upcoming Talk

2017.04.26 Peter Jenks (UC Berkeley)

Writing grammars: Confessions of a syntactician
Rice (2006) highlights the pitfalls of including both too much and too little formal theory into the process of grammar-writing. I describe some of the challenges I am grappling with in writing a descriptive grammar of Moro (Kordofanian:Sudan) as someone who also does research in formal syntax and semantics. I describe a metric which has guided my decision making: reference grammars should be both helpful and accessible to as many kinds of linguists as possible, a basic utilitarian measure of value. I contend that part of being helpful to those interested in formal syntax and semantics in particular is including data that are not typically included in descriptive grammars but which bear on crucial questions of syntactic analysis. I focus on two test cases: descriptions of quantifiers and descriptions of raising and control predicates.

Preliminary Schedule (Spring 2017)

2017.01.18 Konrad Rybka (UC Berkeley)

2017.01.25 Andrew Garrett (UC Berkeley)

Writing Grammars: Lesson 1

The eponymous lesson(s), for myself, emerge from problems I've faced in grammar writing after struggling to think of the appropriate organization of linguistic analysis in a medium used by typologists, theoreticians, text scholars, and language teachers (and learners). The major issue I want to talk about is "emic" vs. "etic" approaches to derivational morphological, syntactic, and semantic description. If time permits, I'll also talk about paradigmatic vs. item-and-arrangement descriptions of inflectional morphology. No real solutions should be expected.

2017.02.02 Rosa Vallejos (University of New Mexico) Note: this meeting will take place at a special time, namely on Thursday 02/02 from 12:30 - 2:00.

2017.02.08 Jed Pizarro (UCSC)

Out of the lab and into the field: An auditory masked priming study in Dabaw Bisaya spoken word recognition

Most of our knowledge about how humans process language comes from psycholinguistic data on Western languages (and a handful of East Asian languages). Traditionally, these behavioral data are collected in laboratory settings. Recently, there’s been a growing interest in “field psycholinguistics.” Researchers have started taking experiments out of the lab and applying laboratory methodologies to lesser studied languages. This movement, while promising, requires for existing methodologies to be adapted so that they become simpler, more portable, and less culturally entrenched. In this talk, I discuss one of the experiments that I conducted in Davao City, Philippines in January 2017. 

The study had two goals: one empirical and one methodological. The first goal was to examine the effect of nasal substitution (NS), if any, on lexical access. NS is a pan-Western Malayo-Polynesian phenomenon where a prefix-final nasal fuses with a stem-initial obstruent (Zuraw, 2010). Does NS obscure the recognition of the verbal stem and thus, impede word recognition in real-time? Framed differently, can the hearer “undo” phonology quickly to access the putative parts of a complex word? I investigated this question with respect to NS in Dabaw Bisaya, a variety of Cebuano (Austronesian) spoken in Davao City. 

The second goal was to explore auditory masked priming (AMP; Ussishkin et al., 2015), the auditory analog of visual masked priming (VMP; Forster & Davis, 1984), the technique commonly used to investigate morphological decomposition and more generally, lexical processing. The study sought to validate whether the robust priming effects found in VMP can be replicated using AMP. If validated, its implications are far reaching: we could use it to investigate morphological decomposition and lexical processing in child language, and languages with no standard orthography, those that are predominantly unwritten, and use such data to bear on theories of lexical access. 

2017.02.15 Ruth Rouvier, Amy Avishay, and Frederica Priyanto (UC Berkeley)

Language Documentation and Revitalization, and Impacts on Child Learners

Slides for this talk can be accessed here.

Documentation of endangered languages provides critical linguistic resources to efforts to support endangered language (re)learning in community and institutional contexts. However, to date there has been very little research focused on exactly how, why, and to what extent documentation benefits these efforts, and subsequently how to achieve better cohesion between documentation practices and revitalization goals. Similarly, there is little documentation of child language acquisition of endangered languages, important for understanding these unique language learning situations and for shedding light on language change in endangerment contexts. There are also questions regarding extra-linguistic benefits that may result from efforts to maintain or re-introduce an endangered language into the community using language documentation resources. 

We discuss the preliminary findings of an interdisciplinary working group of researchers and practitioners who met in October 2016 to address the role of language documentation in language revitalization activities, and the impact of these activities on the crucial population of learners ages 0-5. The group is made up of researchers and educators from a variety of disciplines and communities, including descriptive and applied linguistics, public health, pedagogy, psychology, education policy, and child development. This group came together with the goal to build a network capable of moving forward with research on language endangerment and revitalization across disciplinary divides. We considered the impacts of language documentation and revitalization from an interdisciplinary perspective, including increased fluency and number of speakers; improved mental and physical health; improved academic outcomes (both in assessment scores and retention rates); sense of cultural identity and leadership skills; and stronger community and family connections. We reviewed existing research on these issues, and identified areas needing further exploration as well as methodological concerns for future research.

2017.02.22 GAIL Meeting Note: this meeting will take place at a special time during the evening on 02/22.

2017.03.01 Hannah Sande (UC Berkeley)

Challenges in developing an orthography

As fieldworkers we are often confronted with the task of developing an orthographic system for the languages we work on. In this talk I detail the challenges that have arisen in working with the Guébie community to develop an orthography for use in community language materials. These challenges range from the representation of tone to differences of opinion between linguists and community members. I present the difficulties I have faced in developing a Guébie orthography in collaboration with community members, and I leave for discussion a number of unsolved problems in the Guébie orthography project, and in orthographic development in general.

2017.03.08 Bernat Bardagil-Mas (UC Berkeley)

How to manage blind spot dangers in language description

This talk follows up on previous forums this semester on language description and grammar writing. In the last months I have become increasingly aware of the internal and external biases that a field linguist can be an unaware victim of. I would like to create a discussion about how certain conceptual and theoretical assumptions that operate under our radar can influence both our description of a language and the process of committing to a particular grammatical description. The discussion will kick off with approaches to grammatical relations by looking at particular cases of Amazonian languages, including Katukina-Kanamari, Trumai and Panará. The usual disclaimer holds: No real solutions should be expected.

2017.03.15 Chris Beier (UC Berkeley)

Fun with homoFones!  The challenges of homophony in documenting Iquito (iqu; Zaparoan; Peruvian Amazonia)
A significant challenge in developing a complete and accurate description of Iquito (Ikíitu) grammar, as well as in faithfully transcribing, interlinearizing, and interpreting recorded Iquito texts, is posed by sets of homophonous and near-homophonous morphemes (and syllables) in the language, many of which are high-frequency in connected speech. In this talk, first I will give a bird's-eye view of the types and distributions of these elements in Iquito. Then I will delve into few illustrative cases  -- including 8 kinds of "na(a)", 6 kinds of "ma(a)" and the cluster ki(i)~kia(a)~kiyaa -- with the unifying goal of sharing some methodological perspectives and tools that have helped me better understand this always fascinating, sometimes inscrutable, phenomenon.

2017.03.22 Gabriela Caballero (UCSD)

Writing grammars: tone and morphological structure in Choguita Rarámuri (Uto-Aztecan)

In this talk I report on some challenges that arise in writing a grammar of Choguita Rarámuri (CR), a tonal Uto-Aztecan language of Northern Mexico. CR exhibits a three-way lexical tonal contrast that plays a role in the morphological organization of the language: tone has a paradigmatic role, both in terms of its association to other aspects of word prosody and in an autonomously morphological way. Given that tone is also involved in the intonational encoding in this language,characterizing the ways in which tone participates in morphological structure vs. the lexical phonology and intonation raises questions as to how to structure its description: what is the best way of characterizing these complex interactions between different modules of the grammar? Are there specific formats that are preferable in terms of potential audiences (typologists, Uto-Aztecanists or language learners)? This talk also addresses the ways in which the reference grammar of CR will be linked to a documentary corpus in order to provide access to data on variation in prosodic structure.

2017.03.29 No Meeting - Spring Break

2017.04.05 Robert Geary (Elem Indian Colony)

Robert Geary is a member of the Elem Indian Colony of Pomo Indians of Lake County, California. He is the tribal community language instructor, leader of the traditional ceremonial roundhouse, and the Founder and President of the Clear Lake Pomo Cultural Preservation Foundation. In 2001, Robert began learning and documenting the Xai-tsnoo language from the last two Southeastern Pomo speakers, Loretta Kelsey and Elvira (Patch) Wilder. Robert and Ms. Kelsey participated in the Master Apprentice Program under the Advocates for Indigenous California Language Survival and were consultants for a field methods class at UC Berkeley. Robert recently returned to college to pursue a degree in Native American Studies to help preserve his native culture and language.  Robert will discuss his experience working with the last fluent Xai-tsnoo speakers, linguists, and archival materials to revitalize his heritage language.

2017.04.12 Brook Lillehaugen (Haverford College) Note: this meeting will be held through GAIL and will take place at a different time during the evening on 04/12.

Archives, morphological analysis, and XML encoding: interdisciplinary methods in the creation of an online corpus of endangered language texts

Zapotec (Otomanguean), an indigenous language family of Oaxaca, Mexico, has a long record of alphabetic texts, the earliest dating to 1565 (Oudijk 2008). Interpreting these colonial documents can be difficult because of the challenges of early orthography, grammar, and printing conventions.  In this talk I describe how linguistic knowledge and interdisciplinary methods can be brought to bear on this corpus of texts in order to facilitate the use of the historical material by non-specialists. I describe the project Ticha: a digital text explorer for Colonial Zapotec (Lillehaugen et al. 2015), which allows users to explore multiple layers of these texts, including images of the original documents, transcriptions, translations into English and modern Spanish, linguistic analysis, and commentary. 

Access to these texts has consequences outside of the field of linguistics, including history, in which indigenous Zapotec writings have been largely ignored, apart from the work of a small handful of ethnohistorians.  Moreover, this historical corpus of texts exists alongside modern Zapotec speaking communities, many of whom do not know that these texts exist.  There is great value in having a demonstrable written history, especially to speakers of a language that is marginalized. For speakers of Zapotec who are resisting language loss and discrimination, the fact that there are historical written texts can serve as a point of valorization, and the texts themselves can be important instruments for language revitalization. 

2017.04.19 Inge Genee (Lethbridge & Berkeley)

What is a "userfriendly" dictionary anyway?

In this talk I will report on a different kind of fieldwork relating to dictionary-making: not the perhaps more familiar kind where the linguist works with speakers to document the language and create a dictionary, but a secondary kind where the linguist works with community members to turn an existing dictionary made in the traditional way into a resource that can be used in local revitalization efforts.

I will first introduce the Blackfoot Language Resources and Digital Dictionary project (http://blackfoot.atlas-ling.ca), which I direct at the University of Lethbridge. This project began its life as a digital repatriation project with the back-digitization of an existing print dictionary. The next phase of the project works toward turning it into a digital resource that is helpful in revitalization work, in particular in Blackfoot language classes in local schools. The dictionary is part of a site that also includes grammar pages, a story archive and other resource pages and can be extended with other elements as it develops, such as lessons and games.

I will then present for discussion a few issues we are currently working through and which have not found a final solution. All these issues relate to the dictionary itself and our efforts to make it "userfriendly". Some questions include: Userfriendly for who? How best to handle issues around orthography? How much cultural and other encyclopedic information to include? How to prioritize the addition of media such as images, audio and video (especially in relation to limitations of funding and time)? How much and what kind of grammatical information to include? How to handle dialectal and generational variation? I will show what we have done so far, what we think already works pretty well, and what still needs further consultation with users.

2017.04.26 Peter Jenks (Berkeley)

2017.05.03 Discussion of summer fieldwork projects

2017.05.10 Terry Regier (UC Berkeley)

Previous Talks (Fall 2016)

2016.11.29: Myriam Lapierre (UC Berkeley)

The Role of Theory in Language Documentation: The Panará Project

In this talk we discuss the connection between language documentation, language description and linguistic theory, and we put forward the advantages of rethinking traditional approaches to language documentation into a process with a higher degree of involvement of theoretical linguistics.

We then discuss one such example in the authors’ ongoing linguistic work on Panará (Jê) in Central Brazil to document the language and develop materials for use by the community. A digital dictionary and an orthography reform are the concrete materialization of the Panará community’s desire and request. Developing a functional orthography is the first step necessary to text transcription, and consultation with native speaker consultants naturally informs the phonological description. In addition, recorded and transcribed texts provide documentation materials on Panará language and culture, which at the same time feed the digital dictionary with lexical entries, contexts of use and recordings. The linguistic analysis necessary to articulate the various parts of the Panará documentation project is at the same time directly tied to the authors’ own research interests.

The way in which the documentary, descriptive and theoretical elements are combined is fine-tuned to the specific needs and interests of both the speaker community and the researchers involved in the project, resulting in a process with a big community participation that still meets the scientific goals of the linguist's research.

Ruth Kramer (Georgetown University) NOTE THIS MEETING IS ON MONDAY 2016.11.21 9:00-10:30AM

In Amharic, a bare noun can be interpreted as referring to one or more entities.  In this talk, I investigate the properties of these number-neutral nouns and argue that, rather than being (pseudo)incorporated, they are DPs that lack a NumP projection.  I then take a close look at the individual variation in the presence/properties of number-neutral nouns across speakers and speculate on why this variation occurs.  I close with some reflections on doing fieldwork on Amharic and Ethiopian languages in general.

2016.11.15: Kelsey Sasaki (UC Santa Cruz)

Predicate Initiality in Hawai'i Creole

This work in progress investigates predicate-initiality and pronoun-doubling in Hawai'i Creole (HC). Initially documented by Sakoda and Siegel (2003), adjectival predicates can appear sentence-initially in simple declarative contexts. In my recent fieldwork, I found that only certain adjectival predicates can participate in the construction:

(1) Context: You are describing your friend Kimo/the house up the street to me.

     a. Smat (ah) Kimo/him. 'Kimo/He is smart.'

     b. *Yellow (ah) da house.  'The house is yellow.'

This work investigates (i) the semantic distinctions between the predicates that can be initial and those that cannot; (ii) diachronic analyses of the phenomenon, focusing on Japanese and Hawaiian as possible sources, and (iii) the phenomenon's relation to right periphery pronoun doubling in HC, which also occurs in regional Englishes. I discuss challenges that emerged in elicitation, including working with monolingual HC speakers and difficulties eliciting HC at all from HC/English bilinguals. I also seek feedback on designing a database for both the general public and linguists, developing HC learning materials, and filling out the HC predicate-initiality paradigm.

2016.11.08: No Meeting (FForum members are invited to attend GAIL on the following Wednesday evening 2016.11.16)
Vera Gribanova (Stanford University)

Uzbek verbal and non-verbal predicate formation

2016.10.25: Andrew Garrett, Erik Maier, Zachary O'Hagan, and Ronald Sprouse (UC Berkeley)

Archiving your language materials in the California Language Archive

2016.10.18: Rose-Marie Déchaine (Univerisity of British Colombia)

Embracing multi-functionality: a case study of Bantu noun-classes 
Numerous methodological issues arise when collecting data in domains where the target items are multi-functional. Although this is difficult terrain to navigate, it is well worth the effort in terms of the rewards gained relative to enriched descriptive, observational, and explanatory adequacy. This case study of multi-functionality will be based on Bantu noun-class morphology, drawing on data from Shona (southern Bantu) and Nata (eastern Bantu). Bantu noun-class prefixes have numerous functions including (this is not an exhaustive list):
(i) coding the canonical number and gender of underived nouns;
(ii) coding the contrast between underived count and mass nouns;
(iii) deriving de-verbal nouns, where they code the contrast between agent, theme, instrument, and event nominals;
(iv) contributing evaluative force, as when a noun-class prefix combines with a non-canonical noun;
(v) honorific marking, a special case of evaluative force, which also involves the combination of a noun-class prefix with a non-canonical noun.
The presentation has two goals. First, to discuss and get feedback on the most effective way of gathering a complete and representative data set of the different contexts where one encounters multi-functional noun-class prefixes. Second, to discuss the implications of the multi-functionality of noun-class morphology, as most treatments — whether phonological, morphological, syntactic, or semantic — do not take this into account in either the description or analysis of noun-classes. In particular, the multi-functionality of noun-class prefixes indicates that:
• their phonological behavior can be used to investigate the activity of  “minimal word size” constraints, as well as to investigate allomorphy across word-classes;
• their morphological status — especially as regards the divide between inflection and derivation — is less deterministic than most current accounts countenance;
• their organization both confirms and calls into question widely accepted notions of paradigm structure; 
• their syntactic deployment is much more flexible than most current accounts countenance;
• their semantic denotations are much more abstract than most current accounts countenance.
More broadly, though discussion will focus on Bantu N-class prefixes, the methodological problem of controlling for multi-functionality when gathering data presents itself in numerous domains. Consequently, in the interests of developing more granular descriptions, the challenges raised by systemic and pervasive multi-functionality is of vital interest to all fieldworkers. The presentation will close with general comments on the diagnostic tool-kit needed to detect, and play with, multi-functionality.

2016.10.11: Amalia Skilton (UC Berkeley)

Cushillococha Ticuna in a Canonical Typology of noun class
This talk describes for the first time the typologically unusual noun class (gender) system of the Cushillococha dialect of Ticuna, an isolate spoken in northwestern Amazonia. Corbett's (1991) typological survey of noun class presupposes that all noun class systems share two features: noun class (a) exhaustively partitions the lexicon and (b) is an invariant feature of lexical entries which cannot be manipulated to semantic or pragmatic effect. Neither of these generalizations, I show, holds in Ticuna. A large proportion of nouns are underspecified for class, and speakers systematically manipulate noun class agreement to convey both truth-conditional meanings, such as non-specificity, and non-truth-conditional ones, such as social deixis. This is troublesome for our understandings both of noun class -- can this analytic concept stretch to cover cases where agreement is meaningful? -- and of social deixis, which is often understood to hang on the conventional implicatures of specific lexical items (Levinson 2004, Potts 2005). I also comment on methods for successful elicitation of 'optional' agreement and information-structurally sensitive lexical items.

2016.10.04: Eric Zyman (UC Santa Cruz)

P'urhepecha Negative Shift and Cyclic Linearization

Janitzio P'urhepecha (island of Janitzio, Lake Pátzcuaro, Michoacán, Mexico) is a VO language, but negative objects (and negative nonsubjects more broadly) obligatorily appear to the left of the verb. On the basis of new field data, I explore the properties of this phenomenon (negative shift) and note a constituent-order paradox: a negative object must precede the verb; the verb can raise very high (to Mood); but a negative object cannot raise very high. I present a solution to this paradox in terms of cyclic linearization (Fox & Pesetsky 2005, Engels 2012, a.o.). On the methodology side, I discuss some challenges I have faced in investigating this phenomenon, including prescriptivism and interspeaker variation in the form of the negative phrases.

2016.09.27: Ethelbert Emmanuel Kari (University of Botswana)

On the Status of Subject Markers in African Languages

This paper discusses the status of subject markers in African languages, noting that although the grammatical functions of these morphemes are fairly clear, their status as affixes or clitics is not clear. Consequently, subject markers have been analyzed as affixes in some languages and as clitics in others, or even as affixes and clitics in the same language by different researchers working on the language. The paper adopts a descriptive approach. It highlights the fact that in some African languages, subject markers that were previously analyzed as morphological affixes have now been reanalyzed as clitics. It notes that subject markers (and object) markers in African languages appear to be more of clitics than morphological affixes. Given the analysis and reanalysis of subject markers as morphological affixes and/or clitics in the languages discussed herein, the paper recommends that researchers working on African languages having subject markers should look beyond some general or cross-linguistic diagnostic features and possibly combine such cross-linguistic criteria with language-internal pieces of evidence in establishing the status of these bound morphemes as genuine morphological affixes or as syntactic clitics.

2016.09.20: Jason Ostrove (UC Santa Cruz)

Scottish Gaelic

Much work has indicated that morphological (m-) case assignment is formally divorced from nominal licensing (Marantz 1991, Bobaljik 2008). The picture which emerges is that morphological case is assigned post-syntactically. Furthermore, recent work has indicated that other morphological processes like allomorphy selection and phi-feature valuation may be sensitive to linear order (Embick 2010, Arregi & Nevins 2012, Bhatia et. al. 2009, Bhatt & Walkow 2013). This talk explores the possibility that m-case is likewise sensitive to linear order (Adger 2000). I present novel Scottish Gaelic data collected in August, 2016 to argue that an adequate understand of these requires m-case assignment algorithms be sensitive to linear information. These data come from prepositional ("dative") case assignment, which is subject to tremendous cross-dialectal and cross-generational variation, requiring us to likewise confront questions of language change and dialectal landscapes.

2016.09.13: Virginia Dawson (UC Berkeley)

Investigating Indefinites
Cross-linguistic data on indefinite pronouns and determiners is important both for typological studies, and semantic and syntactic theory. In this talk, I discuss what to look out for when documenting a language's indefinites, along with how to collect that information. The discussion is organized around my investigation of indefinite pronouns in Tiwa, a Tibeto-Burman language of Northeast India. There will be room in the talk to bring up data you have/are encountering, so please feel free to bring any along.

 No Meeting
2016.08.30: Welcome back!