Storyboards for Semantic Fieldwork: Looking at Ndebele language data from the Twin Dilemma storyboard

Through semantic fieldwork, linguists can elicit information about meaning in language by interacting with native speakers (Bochnak & Matthewson, 2020). The main aim is to establish facts about meaning of utterances—these are typically subtle and context-dependent (Matthewson, 2004). One method for data collection is through the use of storyboards which consist of picture panels that are matched to a story told in the contact language. The native speaker then retells the story in their language (using the pictures as prompts), meaning elicitation is more natural and spontaneous, with little interference from the contact language (Matthewson, 2011).

 

Language documentation is vital for understanding how different languages construct and encode meaning. Through researching and analysing the syntax and semantics of languages (especially those under-researched/-documented), we gain a greater understanding of what is possible within languages, allowing us to compare and contrast systems (and document endangered languages). Translations, language grammars and dictionaries are inadequate for data collection; they do not provide direct evidence about meaning, nor do they return negative evidence (that is, evidence about what is not possible in a language). In contrast, linguists can use different storyboards that target specific language phenomena to gain a greater understanding of a language.

 

Last year (penultimate year of UG Linguistics programme), we created a storyboard to target comparison constructions in order to provide an initial set of data for how a language encodes the greater-than relation (Marilyn is older than Verna, for example). This storyboard follows a father learning to tell his twins apart, as their mother explains the twins’ differences. Additional elicitation

tasks, such as acceptability tasks, can be used to return negative evidence based on initial data collected from storyboards. To test our storyboard, we used it with a native speaker of Ndebele (Bantu; Zimbabwe) to collect data on the grammar of comparison, informing the syntactic and semantic analysis of the comparative and related constructions in the language. See Hohaus and Bochnak (2020) for a recent overview of the cross-linguistic comparison construction research.

 

Looking at the data elicited with the storyboard with an Ndebele speaker, we found that it was a good starting point for collecting a wide range of comparison constructions, allowing us to form an initial hypothesis as to how Ndebele encodes comparison. Follow-up elicitation with individual panels confirmed the dual-strategy of Ndebele when encoding greater-than relations: an EXCEED-type

comparative (1a), and a locative comparative (1b).

 

(1) Marilyn is older than Verna

 

(a) U Marilyn um-dala u-kwedlul-a u Verna.

1 Marilyn 1-old 1SM-exceed-FV 1 Verna

(Lit.) ‘Marilyn is old, exceeding Verna.’

(b) U Marilyn um-dala ku-la u Verna.

1 Marilyn 1-old LOC-1.DEM 1 Verna

(Lit.) ‘Marilyn is old on Verna.’

 

 

Storyboards can be used in semantic fieldwork to collect language data in a natural, spontaneous manner. Using our storyboard can provide an initial set of comparison constructions in a specific language, offering an insight into the semantic analysis and strategy that a language uses for comparison.