Iconicity and Verb Agreement

A Corpus-Based Syntactic Analysis of German Sign Language

Marloes Oomen

• Éditeur: De Gruyter Mouton
• Année: 2021
• ISBN:
  9783110742848
• Édition: 1
• Sujet: Languages & Linguistics
• Sous-sujet: Sign Language

In many sign languages around the world, some verbs express grammatical agreement, while many others do not. Curiously, there is a remarkable degree of semantic overlap across sign languages between verbs that do and do not possess agreement properties.
This book scrutinizes the interaction between semantic and morphosyntactic structure in verb constructions in German Sign Language (DGS). Naturalistic dialogues from the DGS Corpus form the primary data source. It is shown that certain semantic properties, also known to govern transitivity marking in spoken languages, are predictive of verb type in DGS, where systematic iconic mappings play a mediating role. The results enable the formulation of cross-linguistic predictions about the interplay between verb semantics and verb type in sign languages. An analysis of the morphosyntactic properties of different verb types leads up to the conclusion that even 'plain' verbs agree with their arguments, where iconicity again plays a crucial role.
The findings motivate a unified syntactic analysis in terms of agreement of constructions with verbs of all types, thus offering a novel solution to the typological puzzle that supposedly only a subset of verbs agree in DGS and other sign languages.

Part I: Background

1 Introduction

Almost all sign languages include a subset of verbs that can mark their arguments by modifying their path movement in space, such that it starts at the locus associated with the subject, and ends at the locus associated with the object. Verbs of this type are called agreement or indicating verbs, depending on one’s theoretical approach, and will be referred to as ‘agreement verbs’ throughout this book. Many other verbs cannot be modified in this way: some have a fixed articulation on the body, while others are articulated in space but lack the path movement that characterizes agreement verbs. In addition, there are verbs – generally referred to as spatial verbs – which possess a path movement similar to agreement verbs but agree with locations instead of referents.

Interestingly, verb type membership is known to be at least partially semantically grounded. For instance, a verb denoting the meaning ‘answer’ is a good candidate for an agreement verb. An answering event involves two event participants and a form of (metaphorical) transfer between them; a path movement may fittingly represent such a relation. A verb denoting the meaning ‘feel’, on the other hand, is more likely to be articulated on the signer’s body as a way of reflecting a body-internal experience.

The verb-type system as described above has intrigued many sign linguists over the years, with agreement verbs in particular having received much attention. A number of key questions keep resurfacing: How come that only a subset of verbs can express agreement – and not just with one, but with two arguments? What is the precise relation between a verb’s semantics and its agreement properties? Why do so many sign languages share the same tripartite verb classification? And should the mechanism by which agreement verbs mark their arguments, i.e. through modification of their path movement, be analyzed as proper grammatical agreement or not?

This book provides answers to all of these questions for one specific sign language, namely German Sign Language (DGS). While not being the first to attempt to do so, it differs from many others within this realm of research in devoting equal attention to every verb type. This approach eventually leads to an integrative theoretical analysis, couched within Generative Grammar, that accounts for the syntactic structure of constructions with different verb types in a single unified model. Only spatial verbs are shown to display behavior that is different to such a degree that it warrants an alternative analysis, which will be laid out in a separate chapter.

An important theme throughout this book is the interplay between iconicity (direct form-meaning correspondences) and lexicon and grammar. For instance, the observation that the verb type system in sign languages is semantically grounded can be partially attributed to iconicity, an idea which is further developed in Part II of this book. As will be demonstrated in Part III and theoretically accounted for in Part IV of this monograph, iconicity effects may even reach all the way into the grammar of DGS. It is important to point out that such effects can be accounted for in formal terms without having to directly appeal to iconicity in the formal structure itself, and this is indeed a guiding principle for the theoretical analyses laid out in this book.

This chapter sets the stage for the rest of the book and is organized as follows. Section 1.1 introduces a number of relevant linguistic properties and concepts, in particular pertaining to the use and expression of verbs and their arguments in sign languages. The concept of iconicity is discussed separately in Section 1.2. Section 1.3 provides some background to the historical and sociolinguistic characteristics of German Sign Language, the language of study. The research reported on in this book is based primarily on the analysis of naturalistic corpus data; in Section 1.4, I discuss the advantages and limitations of such corpus-based research. The goals of the investigation are outlined in Section 1.5. Finally, Section 1.6 presents a how-to guide for reading this book.

1.1 Verbs and arguments in sign languages

Section 1.1.1 introduces a number of general properties of sign language structure (readers who are well-versed in the field of sign language linguistics may skip this section). Subsequent sections discuss person marking (Section 1.1.2), verb classification (Section 1.1.3), agreement auxiliaries (Section 1.1.4), null arguments (Section 1.1.5), and agent-backgrounding (Section 1.1.6).

First, a general remark. Although every sign language has its own grammatical structure, there are many properties that are shared or common across sign languages, and this also applies to the phenomena under discussion in this book. Some such similarities may reflect modality effects, but there are other possible explanations. For instance, shared linguistic properties that are also attested in abundance across spoken languages may represent genuine language universals. Another contributor to cross-linguistic similarity among sign languages is their relatively young age: there are no known sign languages older than 300 years (Woll, Sutton-Spence & Frances 2001), and many are much younger than that. Indeed, sign languages have been observed to display striking similarities to young creole languages, in particular in their course of development and acquisition (Fischer 1978; Gee & Goodhart 1988; Aronoff, Meir & Sandler 2005). There are also commonalities in language structure, such as in the prevalence of rich aspectual systems, the frequent lack of prepositions to introduce oblique case, and a general heavy reliance on prosodic cues for expressing particular syntactic relations (Fischer 1978; Gee & Goodhart 1988). Whenever there is a high degree of similarity across sign languages within a particular grammatical domain, it is important to establish what the root cause(s) of the commonalities are. An adequate explanation enables the formulation of cross-linguistically valid predictions and generaliza...