LINGUIST List 27.4551

Tue Nov 08 2016

Review: Discipline of Ling; Semantics; Translation: Matthewson, Bochnak (2015)

Editor for this issue: Clare Harshey <clarelinguistlist.org>


Date: 04-Apr-2016
From: Jesus Villalpando-Quiñonez <jesus.villalpandocolorado.edu>
Subject: Methodologies in Semantic Fieldwork
E-mail this message to a friend

Discuss this message


Book announced at http://linguistlist.org/issues/26/26-4638.html

EDITOR: M. Ryan Bochnak
EDITOR: Lisa Matthewson
TITLE: Methodologies in Semantic Fieldwork
PUBLISHER: Oxford University Press
YEAR: 2015

REVIEWER: Jesus Villalpando-Quiñonez, University of Colorado at Boulder

Reviews Editor: Helen Aristar-Dry

SUMMARY

M. Ryan Bochnack and Lisa Matthewson’s “Methodologies in Semantic Fieldwork” constitutes a full discussion on the recent advances on semantics, especially in underdescribed languages. The book is intended to fill a gap in methodology that has not been addressed before, except for the pioneering paper “On the Methodology on Semantic Fieldwork” by Matthewson (2004).

The book is divided into three main parts. The first part “General Overview of Elicitation Techniques” presents general background information and its relevance to the available linguistic literature.

The second part is dedicated to “Techniques for Particular Semantic Domains”. This section presents a reevaluation of the methodologies in order to better fit the needs of semantic fieldwork, as well as a description of new methodologies that have been proven fruitful for fieldwork semanticists.

Finally, the third part, “Using Language-Internal Evidence to Guide Semantic Fieldwork,” discusses how a theory-based exploration in the field and semantic research provide each other with feedback that brings to light new findings.

The next subsections are dedicated to reviewing the book’s chapters individually.

PART I. GENERAL OVERVIEW OF ELICITATION TECHNIQUES

The first chapter (“A Practical Epistemology for Semantic Elicitation in the Field and Elsewhere” by Jürgen Bohnemeyer) focuses on the question of whether research on semantics should rely on methodologies similar to those used for obtaining data in other subfields, given the assumption that meaning is not directly observable. The main concern for semanticists is what Matthewson (2004) dubbed “relativist agnosticism” or “the widespread assumption that it is impossible to study the semantics of languages the researcher does not speak, or at least does not speak native-like”. In her pioneering paper, Matthewson (2004) pointed out that this was a fallacy.

One of the main questions that the paper addresses concerns how a researcher can gather linguistic evidence with observable properties. According to Bohnemeyer, the collection of linguistic data involves three components: a stimulus, a task, and a response. Based on these components, the author classifies eliciting techniques into seven types: each type is illustrated by examples from Bohnemeyer’s field work on Yucatec Maya.

Finally, the chapter claims that studying semantics in the field or in a lab must not rely on the researcher’s native speaker intuitions but on empirical semantics based on observations of the communicative behavior.

In the second chapter (“The Problem with No-Nonsense Elicitation Plans for Semantic Fieldwork”, by Meagan Louie) encompasses paradigmatic elicitation, a topic well-loved by linguists. Louie centers the discussion on what she considers to be the biggest problem related to paradigmatic elicitation: boredom. In her own words, “[…] while a linguist, as an academic, may be trained to endure situations with advanced levels of boringness, this is rarely the case for their language consultants” (p. 55). This is supported by a review of different elicitation plans designed to systematically elicit specific domains in Blackfoot (Algonquian). All of these plans reveal two characteristics of paradigmatic elicitation plans: either they are mentally straining, too transparent, or both. Louie’s alternative method suggests linking the target utterances and relevant contexts to an overarching storyline.

Section Three of the chapter explains what overarching storylines are, how to come up with outlines, and more importantly, how to interpret and test the findings with this method. In the end, the author lists issues regarding the overarching storyline-style methodology, like taboo words, time invested in creating storyboards and unexpected stylistic choices. Louie concludes that despite the time-consuming first stage, this method gives better results than traditional paradigmatic elicitation, especially over the long term.

PART II. TECHNIQUES FOR PARTICULAR SEMANTIC DOMAINS

The third chapter (“Documenting and Classifying Aspectual Classes Across Languages” by Leora Bar-el) points out that aspectual classes have not been addressed in the literature due to the assumption of their universality, a frequent though unsupported cross-linguistic claim that could be traced back to Vendler’s (1957) Aktionsart. However, semantic research centered on aspectual classes has shown that language-particular aspectual classes may not have been accounted for in the past because of over-applying tests designed for English.

On the other hand, determining aspectual classes in a given language is not as straightforward as previous studies have claimed. The main reasons are: the inapplicability of the already proposed tests in all languages, and intraclass variation.. For instance, in Dëne Sųłiné (Athapaskan), there is only one way of expressing ‘in x time/for x time’ and it is compatible with both activities and accomplishments. Additionally, ‘ʔanast’e’, the closest equivalent to ‘stop/finish’ occurs with both activities and accomplishments. Both tests fail in distinguishing between telic (e.g., accomplishment) and atelic (e.g., activity) predicates in Dëne Sųłiné and many other languages. Section Five presents several examples testing issues and diagnostics with different aspectual classes from a cross-linguistic perspective. Then Bar-el concludes that language-specific tests cannot account for cross-linguistic aspectual classes, even though, “an inventory of building blocks that languages use to construct aspectual classes” (p. 105) may be universal (von Fintel and Matthewson 2008; Tatevosov 2002).

The fourth chapter (“Investigating Gradable Predicates, Comparison, and Degree Constructions in Underrepresented Languages”) by M. Ryan Bochnak and Elizabeth Bogal-Allbritten discusses the recent interest in exploring comparison constructions and gradability. The authors highlight the relative lack of in-depth information about these domains in reference grammars and in cross-linguistic research.

Some of the issues when investigating constructions involving gradable predicates are that they do not represent a uniform class and are highly context-dependent.

The authors present a methodology for eliciting such predicates based on their experience working with Washo (Isolate) and Navajo (Athabaskan), especially with the issue of modifier licensing in these languages. While “very” and “completely” in English track the type of standard, their closest semantic equivalents have different distributions: Navajo “yee’” tracks positive/negative polarity and Washo “šemu” does not track any distinctions.

On the other hand, although exploration of norm-relatedness and crisp judgments in Washo favors the use of targeted visual stimuli, they must be tested separately. Additionally, creating stimuli to test norm-relatedness with evaluative predicates might be difficult, in which case the authors suggest the use of verbal contexts.

In summary, gradable predicates and comparison constructions require the use of a mixed methodology that includes visual and verbal stimuli, in order to avoid context-dependence and norm-relatedness issues.

The fifth chapter (“Targeted Construction Storyboards in Semantic Fieldwork”) by Strang Burton and Lisa Matthewson presents a detailed account of storyboards as an effective way to obtain data in the field.

Compared with traditional semantic elicitation methods which consist of translation, sentence judgments, and elicited productions tasks, storyboards allow the researcher to acquire natural data without verbal interference (cf. Matthewson 2004; Krifka 2011) or other issues frequently related to traditional elicitation. There is the additional advantage of targeting a specific construction or semantic domain.

In Sections Two and Four of the chapter, the storyboard methodology is applied to modal distinctions in three unrelated languages: Gitksan (Tsimshianic), St’at’imcets (Salish), and Blackfoot (Algonquian). Section three is dedicated to describe the methodology itself, how to organize storyboards into coherent, interesting, and target centered narratives. The authors propose a linear order of elements that have worked for them in the past: (i) Introduction to the characters, (ii) Events that lead to the target structure, (iii) Iteration of the targeted narrative context, and (iv) An engaging conclusion with a plot twist. As for creating the storyboards, the authors encourage the researcher to think thoroughly about the context where a specific construction would occur, and then sketch the storyboard.

Finally, in Section Five, the authors tested the naturalness of the output narratives obtained through the storyboard methodology and the narrative elicited. Linguists who were native speakers of Japanese were asked to judge naturalness of both outputs using the following criteria: (i) vocabulary of choice, (ii) intonation patterns, (iii) narrative transitions, and (iv) use of discourse sensitive markers. Their results show that narratives obtained through storyboards were judged to be not only closely similar to spontaneous elicited narrative but also to be even more natural for some speakers.

The sixth chapter (“Reasoning About Equivalence in Semantic Fieldwork”) by Amy Rose Deal, deals with the most basic question of semantic fieldwork: “figuring out how meaning is conveyed in a language” (p.157) that the researcher might not speak natively.

The different methodologies already available for the purposes of semantic fieldwork require the researcher to ratiocinate from the answers consultants provide. Such methodologies rely on two hypotheses: the Equivalent Translations Hypothesis (ETH) and the Equivalent Judgments Hypothesis (EJH). Each of these presupposes a series of conversational principles that are usually already understood (e.g., we both understand the task, we are being cooperative, etc).

On the other hand, Deal highlights an important point: the impossibility of equivalent meanings among languages. Even though all languages are able to express any meaning, it is also the case that a synchronic stage of a certain language may not have a means to express it. For instance, in Nez Perce,, there is no way of expressing a non-epistemic necessity mode lexically, like the English modal ‘must’.Nevertheless, situational inferences may express such modal domain (i.e., Equivalent Implicatures Hypothesis or EIH). In summary, the current hypotheses and methodologies do not always arrive at the best conclusion. However, the reasoning about the data and its interpretation is what is ultimately under revision.

The seventh chapter (“Investigating D in Languages With and Without Articles”) by Carrie Gillon explores the cross-linguistic variation of articles in languages with and without articles. The semantic notion of “definite” is commonly related to articles but does not seem to be an unequivocal characteristic since indefinite articles are frequently found in natural languages. The author provides a characterization of what should be considered an article in function and distribution to show their semantic variation cross-linguistically by using a sample of five unrelated languages: English, Skwxwú7mesh (Salish), Lithuanian (Baltic), Innu-aimun (Algonquian), and Inuttut (Labrador Inuktitut; Eskimo-Aleut).

Articles can be defined from different perspectives. Their function is to “create an argument out of a predicate” (Higginbotham 1985; Szabolsci 1987, 1994; Stowell 1989; Longobardi 1994). Thus they are similar to demonstratives but lack the deictic and independent occurrences. They are syntactically flexible and can occupy different positions within a noun phrase (Gillon 2013, following Epstein 1999; Lyons 1999; Borer 2005). As for their semantics, it is well-known that they vary (Matthewson 1998) to encode not only definiteness but also specificity (e.g., Samoan) and even deictic information, as in Salish languages.

An interesting question raised in the paper relates to what articleless languages do in order to fulfill this apparent grammatical gap. Articleless languages only use bare nouns which are also proven to vary semantically from language to language. However, the author supports from the sample the use of covert articles in three of the articleless languages.

Section Five presents four semantic tests that were applied to both article and articleless languages in order to narrow down semantic variation, while Section Six tests the five languages under study in detail by providing contrastive examples which help narrow down the specific function of articles in each language. The semantic tests involve: (i) definiteness, (ii) scope, (iii) law of contradiction, and (iv) domain of restriction.

Interestingly, definiteness was not found in the three articleless languages of the sample. Despite this, the author highlights the fact that syntactic tests may aid the study of articles and their variations.

PART III. USING LANGUAGE-INTERNAL EVIDENCE TO GUIDE SEMANTIC FIELDWORK

The eighth contribution opens the third subpart of the book. In this chapter “Linguistically Establishing Discourse Context: Two Case Studies from Mayan Languages”, Scott AnderBois and Robert Henderson discuss the necessary context for testing truth values and felicity judgments for elicited sentences. In particular, the authors address the kind of context that should be included when eliciting sentences (e.g., verbal, non-verbal, etc). They also take into account the practical and sociolinguistic concerns regarding data collecting. For those cases where verbal contexts are used, they discuss which language should be utilized, that is, the object language (hereinafter OL) or the language of wider communication (LWC). The authors do not use terms such as ‘meta-language’ or ‘contact language’ so as to avoid some scenarios where the term does not describe the researcher’s position accurately.

Sections Three and Four show how discourse contexts were determined during fieldwork in the languages of the authors’ expertise: Yucatec Mayan and Kaqchikel. Yucatec Mayan reported attitudes use the topic verbal suffix -e’, which occurs in other topic constructions. Native speakers as well as linguists have considered that the occurrence or absence of the suffix is subject to free variation. The authors tested both forms (i.e., with and without -e’) as the answer to a Question Under Discussion (QUD), noticing that one form was preferred over the other depending on the QUD asked. Furthermore, by using Spanish as the LWC, the authors proved that the LWC avoided the target construction during tests since that distinction is not overtly marked in that language.

Unlike Yucatec Mayan, evidence from Kaqchikel shows that the OL is the preferred means to present discourse context (e.g., pluractionality in this particular case). Uses of the pluractional suffix -la’ in Kaqchikel allow cumulative readings that are not always captured by its Spanish translations. Furthermore, speakers of Kaqchikel showed attitudes towards Spanish that led the researchers to conduct the research in the OL.

In summary, by presenting two cases in which either the OL or the LWC were chosen as the language for presenting discourse context, the authors claim that “the key factor in determining which language to use is the ability to control the information in the context.” (p. 230). If the researchers were to disclose the details about the contexts described to the consultants, the results of these linguistic studies would benefit, for instance, by allowing other researchers to replicate the study in other languages or even on the same language.

The ninth chapter, “Semantic Fieldwork on TAM” by Rebecca Cover, presents a proposal for collecting data to use in research on Tense, Aspect and Mood (hereinafter TAM). Cover claims that it is possible to do semantic research in a language that is not the researcher’s mother tongue.

In order to give a comprehensive account of TAM categories or any other, the researcher should apply three independent but complementary methods: (i) one-on-one semantic elicitation,(ii) text collection/analysis, and (iii) participant observation of natural discourse. Other tasks such as translation are unreliable for exploring TAM categories due to the language-specific nature of these categories. Another intervening factor is that translation requires a high level of fluency in the language under investigation or the meta-language. However, that does not mean that translation is forbidden. As Cover claims “both fieldworker and consultant need to learn to talk about the meaning of sentences […] often in ways that ‘ordinary’ speakers would find highly unnatural” (p. 240).

The author shows in detail the utility of the methodology applied to her own fieldwork on Badiaranke (Niger-Congo) and specifically on two TAM categories: the imperfective aspect, and the discontinuous past markers -ako- and -akəd-. Some distinctions may be more reliably obtained through one of the methods than the other. For example, imperfective constructions are more easily found in elicitation than in texts in Badiaranke (cf. Chelliah 2001). On the other hand, meaningful or interesting sentences from a text constitute future batteries for elicitation.

The tenth chapter entitled “Deriving Topic Effects in Kiowa with Semantics and Pragmatics” by Andrew McKenzie discusses the status of dislocated Determiner Phrases (hereinafter DP) in Kiowa. According to McKenzie, DP dislocation in Kiowa has not been addressed from a perspective that accounts for all cases while also explaining motivations for speakers to use it. Notions such as topicality, topic, and topichood applied with a purely syntactic approach are not sufficient to capture all the semantically and pragmatically motivated ‘topic effects’. Section two of the chapter is dedicated to show that lack of explanation in a syntactically-based perspective.

One of the important issues that the paper deals with is how to elicit topics, especially when they are not overtly marked. McKenzie shows that most nuances of the discourse behavior of these dislocated phrases are indeed possible to elicit by testing their acceptability in controlled contexts as well as through direct elicitation.

Sections Three and Four present the formal foundations for the Topic Derivation Hypothesis. The author concludes that DPs in Kiowa are not dislocated because they were signaled as topics but the other way around, that is, dislocated phrases are interpreted as topics because of a pragmatic reason (i.e., disambiguation), favored the dislocation. The paper strongly defends the claim that semantic and pragmatic motivations can trigger syntactic variation.

The eleventh chapter of the book entitled “Reciprocity in Fieldwork and Theory” by Sarah E. Murray is twofold. On one hand, it explores the grammar of the Cheyenne language (Algonquian) in order to determine the function of the morpheme -ahte which is used to express both reflexivity and reciprocity. The second goal is to present some notions of formal semantics that the author considers useful for fieldworkers. Section Two describes the basics of Cheyenne required to better understand the paper as well as the methodology used to gather data.

Cheyenne uses the verbal suffix -ahte to express both reflexivity with intransitive verbs and reciprocity with transitive verbs. The use of the modifier nonámé’tóe, which in other non reflexive/reciprocal contexts means ‘one by one’, makes a sentence unequivocally reciprocal. On the other hand, Murray’s main claim in this chapter has to do with a third possible interpretation of the suffix -ahte that is, a mixed scenario where some participants act reflexively while the rest reciprocally, which is common interpretation in other languages with no distinction between reflexive/reciprocals. Murray determines that Cheyenne’s -ahte is not ambiguous but underspecified for reflexivity and reciprocity in those mixed contexts.

In sum, Murray states that training in formal semantics is rather a useful tool for fieldworkers since it allows the researcher to generate hypotheses from an abstract perspective to later test them in the language of study.

The twelfth and final chapter entitled “Theories of Meaning in the Field: Temporal and Aspectual References” by Rebecca T. Cover and Judith Tonhauser constitutes a comprehensive account of the theoretical and practical issues a semanticist fieldworker has to face, including those such as aspectual and temporal reference. The authors claim that a theory-guided exploration does not always lead to proposing nonexistent categories in the language of study.

Cover and Tonhauser dedicate the second section to an explanation of the neo-Reichenbachian theoretical framework which distinguishes between three time intervals: the evaluation topic, the topic time, and the eventuality time. They address how these three terms interact in a clause, resulting in different temporal or aspectual references.

Section Two presents an important distinction between tense and temporal references, as well as aspect and aspectual references. Then, section three introduces the differences regarding temporal reference in matrix clauses and subordinate clauses.

Finally, Section Four explains the theoretical overlaps that some TAM categories might show in order to then demonstrate how they occur in different languages. The authors also highlight the well-known fact that lexical aspect plays an important role in expressing aspectual references.

In summary, Cover and Tonhauser propose a series of requirements that a fieldwork-based and theoretically informed description of meaning should include.

EVALUATION

A contribution to the literature on methodology like Bochnak and Matthewson (2015) will greatly influence future research on fieldwork semantics. New researchers as well as seasoned scholars will benefit from the findings, detailed experiences and practical solutions proposed in this book. The edited book condenses many of the recurrent ideas regarding how to do semantic fieldwork which have never been discussed with sufficient detail or support.

Even though some of the chapters may describe complex semantic domains, the compilation is relatively accessible for both specialists in semantics and a general audience of linguists. The use of relatively formal terminology varies, but this is an expected result of the book’s being an edited volume covering several subjects. Thus, each chapter explores some of the main topics studied in semantics: from definiteness, tense, aspect, mood, and modality, to practical matters like elicitation and alternative ways for obtaining natural data, as well as targeted data. In other words, although the book itself does not follow the structure of a handbook, it provides detailed information on some semantic domains and the way the authors’ fieldwork methodologies have been proven to be fruitful. However, because the book consists of different proposals of a single, integral methodology, several chapters dealt with similar topics, like tense and aspect.

In summary Bochnak and Mathewson (2015) represents not only an in depth guide to effectively explore semantic domains but also the foundations for future research applying an openly disclosed and systematic methodology.

REFERENCES

Borer, Hagit. 2005. In Name Only. Structuring Sense. Vol. I. Oxford: Oxford University Press.

Gillon, Carrie. 2013. The Semantics of Determiners: Domain Restriction in Skwxwú7mesh. Newcastle Upon Tyne: Cambridge Scholars Publishing.

Higginbotham, James. 1985. On Semantics. Linguistic Inquiry 16: 547–593.

Krifka, M. 2011. Varieties of semantic evidence. In Semantics: An International Handbook of Natural Language Meaning, K. von Heusinger, C. Maienborn, and P. Portner (eds.), 242–267. Berlin: de Gruyter.

Longobardi, Giuseppe. 1994. Reference and proper names: A theory of N-movement in syntax and logical form. Linguistic Inquiry 25: 609–665.

Lyons, Christopher. 1999. Definiteness. Cambridge: Cambridge University Press.
Matthewson, Lisa. 1998. Determiner Systems and Quantificational Strategies: Evidence from Salish. The Hague: Holland Academic Graphics.

Matthewson, Lisa. 2004. On the methodology of semantic fieldwork. International Journal of American Linguistics 70(4): 369–415.

Stowell, Tim. 1989. Subjects, specifiers and X-bar theory. In Alternative Conceptions of Phrase Structure, Mark Baltin and Anthony Kroch (eds.), 232–262. Chicago: The University of Chicago Press.

Szabolcsi, Anna. 1987. Functional categories in the noun phrase. In Approaches to Hungarian, vol. 2, Istvàn Kenesei (ed.), 167–189. Szeged: JATE.

Szabolcsi, Anna. 1994. The noun phrase. In The Syntactic Structure of Hungarian, Syntax and Semantics, vol. 27, Stephen R. Anderson and Katalin F. Kiss (eds.), 179–275. New York: Academic Press.

Tatevosov, Sergej. 2002. The parameter of actionality. Linguistic Typology 6: 317–401.

Vendler, Zeno. 1957. Verbs and times. The Philosophical Review 66 (2): 143–160.

von Fintel, Kai, and Lisa Matthewson. 2008. Universals in semantics. The Linguistic Review 25: 139–201.


ABOUT THE REVIEWER

Jesús Villalpando-Quiñonez is currently a Ph.D. student at the University of Colorado Boulder. His interests are language documentation, grammatical description and diachronics. He has conducted research on Southern Uto-Aztecan (UA) languages like Yaqui, Mayo, Nevome and Tarahumara. His current research explores aspect and tense in Tarahumara narratives.

Page Updated: 08-Nov-2016