First part: Multimodal Language Use
NĂșria Alturo, Ignasi Clemente, Universitat de Barcelona, and LluĂs PayratĂł, Hunter College, City University of New York
Notes for a Multilingual and Multimodal Functional Discourse Grammar
Abstract: In this chapter, we argue that Functional Discourse Grammar (FDG), with any adjustments that may be required, can constitute a useful model to explain grammatical phenomena associated with speakersâ multilectal and multi-modal communicative competence. In particular, we present two working hypotheses of how multilingualism and multimodality may be articulated within a FDG grammatical model: (1) languages known by the speaker provide the language-specific primitives and operators that allow the language mixing and switching operations; and (2) speech and gesture share the same primitive frames and templates, and work together in an integrated manner, in the operations of formulation, encoding and decoding. Our evidence suggests that there is a high degree of integration of language systems (verbal grammars) and modes (verbal and non-verbal), and that the contrast between primitives (which may keep the specificity of the languages and the modes involved) and levels of representation (which are specific of each multilingual and multimodal grammar) is a promising perspective to consider in future research.
Keywords: Functional Discourse Grammar (FDG), primitives, operators, levels of representation, multilingual grammar, multimodal grammar
1Introduction1
Formal linguistics has traditionally viewed grammar as a system with an ideal speaker-hearer, who is a member of a homogenous community, and who exclusively communicates verbally. Such view of grammar has resulted in a focus on speaker competence and in the neglect of multiple aspects of performance. Pragmatics has often become a âgrammatical binâ in which multiple linguistic variation phenomena that are considered speech-related, not systematic enough, and not worthy of attention, are thrown in. However, it is difficult to adhere to this view of grammar for a number of reasons. First, research on multilingualism has documented phenomena such as cross-linguistic transfer and code-mixing and code-switching competences, the development of an interlanguage among second language learners, and the creation of pidgins and creoles. Second, research on multimodal language use has revealed that meaning-making is not an exclusive verbal process, but a process that involves different modalities and communicative means that are used simultaneously with talk; for instance, prosody, facial expressions, and manual gestures.
The goal of this chapter is to show that a pragmatically-based Functional Discourse Grammar (FDG) requires an architecture â understood as a structure of frames, figures, and constituents â that is broad enough to include multimodality, multilingualism, multidialectalism, and even the multilectalism that results from combining different registers. Although the grammatical model proposed by FDG restricts what is considered grammatical to those verbal aspects of a language that have systematic codification (Hengeveld and Mackenzie 2008), it acknowledges that communicative performance is nonetheless heterogeneous and variable: communicative interaction is the result of converging discourse modalities (e.g., verbal, vocal, and gestural), and grammars (e.g., languages, dialects, and registers). Thus, a general theory of communication should be able to explain a speakerâs pragmatic competence, that is, the set of competences that lies between a grammatical competence understood in a strictly Chomskyan sense, and a sociocultural and cognitive communicative competence understood in a wide-ranging sense. We believe that a functional grammar like FDG, with any adjustments that may be required, can constitute a useful model to explain grammatical phenomena associated with speakersâ multilingual and multimodal abilities.
In the sections below, we suggest that FDG may provide an adequate framework to build a model of a multilingual and multimodal grammar. In particular, we introduce two working hypotheses of how multilingualism and multimodality may be articulated within a FDG grammatical model: (1) languages known by the speaker contribute the language-specific primitives that allow the language mixing and switching operations of Formulation, Morphological Encoding, and Phonological Encoding; and (2) while the separation between speech and gesture in grammar is limited to the primitive forms and operators available, speech and gesture share the same primitive frames and templates, and work together in an integrated manner, in the operations of formulation, encoding and decoding. With an exploratory goal, we present and discuss these hypotheses, which will need to be confirmed in subsequent empirical work.
2FDG as a model of grammar and levels of adequacy
FDG has its origins in Dikâs Functional Grammar (Dik 1997a, 1997b). By the beginning of the 21st century, FG scholars began to discuss the limitations of the model, including its criteria of adequacy to psychological, sociocultural and most importantly discourse aspects. The result of those discussions was an advancement of FG and the adoption/development of a new model called Functional Discourse Grammar (FDG).2
2.1FDG main features
FDG is a functional, structural and typologically-based theory of grammar. As a functional grammar, it aims at explaining how the ideas and intentions of individuals are formulated and encoded through the grammar of a particular language. It deals, first of all, with pragmatics, semantics, morphosyntax and phonology in grammar; but it acknowledges that grammar cannot be adequately explained without considering its interaction with the non-grammatical aspects of human communication (Hengeveld and Mackenzie 2008; Keizer 2014, 2015). Thus, FDG is thought as the Grammatical Component of a wider theory of verbal interaction, where it interacts with a Conceptual Component, a Contextual Component, and an Output component. As Figure 1 shows, this is a top-down model, working down from the Speakerâs prelinguistic conceptual information and communicative intention to acoustic, orthographic, or signed output, including systematic gesture in speech and other nonverbal aspects of multimodal discourse. The basic unit of analysis is not the Sentence or the Clause, but the Discourse Act, that is, the unit that expresses the communicative intention. Besides, the model captures the role of discourse context and situational context (physical and social) in the production of a linguistic expression (Alturo et al. 2014; Connolly 2013; Connolly 2014; Cornish 2009; Rijkhoff 2008).
Figure 1: General layout of FDG
FDG is, also, a structural model of grammar. It analyses the linguistic representation of Discourse Acts at all levels of grammar, which allows it to account not only for clauses and sentences, but also for units smaller and larger than the clause; that is, interjections, incomplete sentences and sequences of sentences. There are four levels of grammatical organization: two levels of Formulation (the Interpersonal Level and the Representational Level), and two levels of Encoding (the Morphosyntactic Level and the Phonological Level). Furthermore, the operations of Formulation and Encoding are fed by a number of primitives: a list of possible structures (frames, templates), forms (lexemes, grammatical morphemes, suppletive forms) and operators relevant for each level (interpersonal, representational, morphosyntactic and phonological operators). Primitives play a main role in our proposal, and they will be further commented in sections 4 and 6.
The Interpersonal (IL) and the Representational (RL) levels are the outputs of pragmatic and semantic Formulation. The Morphosyntactic (ML) and the Phonological (PL) levels specify the exact way in which the pragmatic, rhetorical and semantic material is encoded. Each of these four levels is hierarchically organized in various layers. For instance, the structure of the IL is formed by a number of layers: Move, Discourse Act, Illocution, Participant, Communicative Content. These layers, as well as the layers of RL, ML and PL, have the general structure represented in (1), where α represents a variable that is restricted by a head, Ï an operator representing grammatical information, and Ï a modifier providing lexical optional information.
(1)(Ï1 α1: [head] (α1): Ï1 (α1))
A simplified representation of the hierarchical structure of the grammar levels is given in (2), which we explain below. Within a particular level, each layer has its own set of operators, functions, and potential modifiers, which are not considered in the example.
(2)Plou âit rainsâ (Catalan)
Interpersonal Level (IL) | (M1: (A1: (F: DECL (F)) (P1)S (C1: (T1) (C1)) (A1)) (M1)) |
Representational Level (RL) | (p1: (present ep1: (e1: (f1: ploure (f1)) (e1)) (ep1)) (p1)) |
Morphosyntactic Level (ML) | (Le1: (Cl1: (Vp1: (Vw1: ploure-prs.ind.3.sg (Vw1)) (Vp1)) (Cl1)) (Le1)) |
Phonological Level (PL) | (U1: (IP1: (PP1: (PW1: (F1: (S1: /plÉw/ (S1)) (F1)) (PW1)) (PP1)) (IP1)) (U1)) |
The Move (M), at the Interpersonal Level, is considered the minimal free unit of discourse and the largest unit of interaction relevant to grammatical analys...