Fieldwork Forum (FForum)

Ese Ejja (Takanan) is an endangered language spoken by around 1500 people in both the Bolivian and Peruvian lowlands I described in Vuillermet (2012). This Amazonian language displays a rich verbal morphology, especially to encode concepts that belong to the expression of space, as can be observed in (1).

1. Nekisowakiani.

   neki	-sowa	-ki	-ani
   stand	-go.up	-GO.TO.DO	-sit.PRS
   #1	#2	#3	#4

   ‘He goes (there) and stands up (on the rock).’

   In (1), four spatial elements can be identified in the verb form:

   - #1 and #4 are posture verbs (the latter standing for TAM);

   - #2 encodes direction;

   - #3 adds (among others) motion and deixis to the whole predicate.

   This talk is about the #3 morpheme, which conveys highly complex semantics and belongs to the set of 12 ‘associated motion’ morphemes of Ese Ejja (Vuillermet 2012a; 2012b; to appear). This little known verbal category of ‘associated motion’ is used to specify backgrounded motion linked to the event expressed by the main verb it is attached to. The phenomenon was first described for Central Australian languages (Koch 1984; Tunbridge 1988; Wilkins 1991). It is also a phenomenon of the Takanan family and was first identified by Guillaume (2006) who described the 11 ‘associated motion’ suffixes of Cavineña. Guillaume (2011) also suggests that the phenomenon might not have been identified yet for other Amerindian languages because it is sometimes confused with ‘directionals’. However, ‘associated motion’ morphemes differ from these better known directionals by two characteristics: first, their semantics are not restricted to path but include also motion, and, second, they are most generally associated to action verbs:

2. Ka'a-nana-kwe!
   lock-DO&LEAVE-IMP
   ‘Close the door before you leave!’
3. Ekwana taxa-taxa-ña-'ya-(a)ni.
   1PL wash-RED-DO.ARRIVING-TEL-sit.PRS
   ‘When we arrive (at the source), we do the washing.’

The semantic complexity mentioned earlier resides in the fact that ‘associated motion’ morphemes also encode temporal relativity with regard to the verb event, and refer either to prior (MOVE & do, as in (3)), concurrent (do WHILE MOVING), or subsequent motion (do & MOVE, as in (1) and (2)).

In this talk I will (1) discuss the semantic parameters that stand in a systematic opposition, (2) describe their function and highlight why they should not be confused with other spatial morphemes related to space, (3) mention other ‘associated’ features found in other languages. In the end, I will briefly introduce my present postdoctoral project (Fyssen Foundation) for the elaboration of an elicitation kit aimed at facilitating the collection of data about such associated motion morphemes in other Amazonian languages.

HANDOUT

We will describe a system, now being set up, whereby Berkeley students and faculty can “pre-archive” digital field materials as they are created in preparation for permanent archiving with the California Language Archive. A beta version of the system will be demonstrated and feedback will be requested from potential users. We would like the system to be as easy to use as possible, given the constraints imposed by archival best practice.

READING 1 (Hill 2005)

READING 2 (Tedlock 1991)

This presentation will introduce Story-builder, a set of picture cards designed to facilitate creative storytelling in any language. The Story-builder card set consists of ‘action cards’ depicting events and ‘character cards’ depicting people; these cards can be placed in various spatial configurations to create narratable visual stories. By drawing on people's natural proclivity to make up stories, the tool engages consultants and language-learners alike to produce speech that is natural and fluid.

In this session we will look at different ways of adapting the Story-builder deck for use in linguistic fieldwork and language pedagogy. This will involve reflecting upon the potential advantages of visual methods more generally, as well as on their biases and limitations. I hope that you will walk away from this presentation with some ideas for your next elicitation session.

Story-builder is freely distributed under a Creative Commons license. For more information, or to download the cards and User Manual, visit: www.story-builder.ca.

SLIDES

I focus on sonority hierarchy-violating sequences in Choapan Zapotec, an Oto-Manguean language spoken in the Sierra Norte of Oaxaca, Mexico. Word-initial onsets in Choapan Zapotec (CHO) frequently consist of consonant clusters that violate sonority sequencing principles. Furthermore, the only codas allowed in CHO are glides and word-final nasals, demonstrating that outside of onset clusters, there are strict sonority-based restrictions governing Choapan phonotactics.

Word-initial consonant clusters in CHO can be tautomorphemic or derived as the result of verbal morphology; both types of onset clutsers can violate sonority sequencing principles. However, violations of such principles are freer across morpheme boundaries than within the same morpheme, suggesting that morphological considerations interact with the phonotactics of CHO in a complex way. Based on data from Choapan Zapotec, I argue that previously suggested approaches to sonority sequencing principles, especially those which tightly link the phonotactics of codas to onset consonant cluster restrictions, should be revised.

HANDOUT

Proposed criteria for defining nominal compounds cross-linguistically are not universally accepted. Moreover, in some languages nominal compounds share properties with phrases and possessive constructions and are not easily identified in all instances (Aikhenvald 2007; Bauer 2006, 2009; Lieber and Štekauer 2009; Saclise and Vogel 2010). This paper examines the phonological, morphosyntactic, and semantic properties of nominal compounds in Chuxnabán Mixe and argues that defining criteria for compoundhood are best viewed language-specifically. In Chuxnabán Mixe, nominal compounds are best discerned phonologically. In addition, they are distinct morphosyntactically by being treated as a whole for inflection and by showing a fixed order with respect to their parts.

This work further establishes that nominal compounding is a productive word-formation process in Chuxnabán Mixe. While nominal compounding has been noted in Mesoamerican (Campbell et al. 1986) and other Mixean languages (Romero 2010; Ruíz de Bravo Ahuja 1980; Schoenhals 1982; Van Haitsma 1967), there are no studies examining its formal properties nor its productivity, possibly due to the fact that nominal compounding is a lesser studied topic in polysynthetic languages where most information is encoded in verbs.

SLIDES

READING (Si 2011)

With this talk, we present multi-media documentation of elders speaking three dialects of Kwak'wala (Wakashan) with the hope of contributing better detail to the underspecified picture of dialect diversity in the language. Because of the connection between dialect and territory, we find the documentation of traditional ecological knowledge (TEK) to be especially well suited for research on dialect divergence. We use video, audio, and still photography to document the yearly-cycle of resource gathering activities in specific places; videos of these journeys are then presented as elicitation stimuli with pairs of speakers. The process yields meta-dialectal reflection by speakers on lexical differences, while allowing for in-depth linguistic analysis of grammatical and phonological differences between dialects. This work arises at the intersection of community interests with linguistic research; we hope that the methodologies presented here can provide a useful model to others interested in combining documentation with revitalization.

This talk presents results of research conducted primarily with a community of refugees in San Diego, California who speak Maay, a Cushitic language of southern Somalia. Perhaps because of the state of severe social disruption experienced by this refugee population, Maay speakers exhibit an extreme degree of linguistic variation. This is particularly true of aspects of the nominal morphology and phonology, which are the focus of this talk. Maay has previously been described based on work with a speaker in Pittsburgh, Pennsylvania (Paster 2007, Comfort & Paster 2009, Paster 2010) as having two regular plural suffixes, -o and -yal, which in some cases can be used together on the same noun (-o-yal) with no apparent semantic or functional consequences. New data show that while some speakers exhibit the pattern originally described, others judge the different plural forms to be semantically different, but in a way that is not entirely consistent from speaker to speaker (and is in some cases completely contradictory between speakers). This suggests that when faced with a redundant system of plural marking in a somewhat unstable language, speakers have each formed their understanding of the meaning of plural nouns in a way that resulted in different semantic representations of the plurals. Speakers also vary in which plural marker goes with which nouns and the extent to which they accept the -o-yal forms. Again, in some cases the data are contradictory from speaker to speaker: for a given noun, one speaker may accept the -o form but not the -yal form, while another speaker does the exact opposite. This again suggests that different speakers have significantly different mental representations, perhaps in terms of which suffix is the default one and/or which nouns have lexically listed plural forms. Finally, speakers differ in their production of vowel length. In some extreme cases, a particular monosyllabic noun is produced by one speaker with a long vowel in all forms in its paradigm, by another speaker with a long vowel that is shortened when the syllable is closed by the addition of a consonant-initial suffix, and by a third speaker with a short vowel that is lengthened when the syllable is opened by the addition of a vowel-initial suffix. I suggest that, based on inconsistent input data, the speakers have formed their own underlying representations of vowel length, combined with lengthening and/or shortening rules that differ from speaker to speaker. There is no coherent way to describe vowel length in the speech community as a whole, but each individual has a clear and describable system.

HANDOUT

READING

A common problem both to field methods classes and to field working teams is to balance the conflicting needs of disseminating language descriptions while allowing these descriptions to change as a language becomes better understood. These challenges become greater when more and more linguists are engaged in a particular project, with a field methods class being among the most extreme such cases. One solution to this problem is to use wiki software, which is designed to be both collaborative and editable. This talk describes how this year's Linguistics 240 (Field Methods) class used MediaWiki, a platform provided by the Wikimedia Foundation, as a tool for collaborative language description and data organization for our investigation of Sereer, a North Atlantic language spoken in Senegal. We will review the motivations for using the wiki, discuss how it was set up on the server, give an overview of the actual grammar and functions of the wiki, and discuss the advantages and drawbacks of using a wiki in a field methods class.