Volume 47, Issue 4

Functional discourse grammar (FDG) is a theory with a rich descriptive apparatus, facilitating the modeling of language data that takes into account the several tiers of interdependent information that are present in any utterance, namely, interpersonal, representational, syntactic, and expression. The explanatory component of FDG comes through complementary principles of linguistic cause and effect, be they diachronic, synchronic, areal, typological, neurological, and so forth. It is this comprehensiveness that sets it apart from many other theories that tend to predicate- or syntax-centricity. FDG is used to analyze the Biblical Hebrew so-called qatal verb (QV), which is characterized by a high degree of multifunctionality. An FDG analysis of several of its functions is provided and although this analysis is shown to be descriptively robust it is also theoretically problematic in some respects. These broader problems in the description and explanation of QV according to the theory of FDG are considered.

The aim of this article is to describe and classify a number of different forms of English reported speech (or thought), and subsequently to analyze and represent them within the theory of FDG. First, the most prototypical forms of reported speech are discussed (direct and indirect speech); subsequently some less prototypical modes of report (free direct and indirect speech, as well as a number of hybrid constructions) are considered. It is argued that these different constructions vary along two major parameters: direct versus indirect and framed versus free. It is therefore suggested that instead of representing different positions on a scale, these constructions are better regarded as occupying different slots in a matrix. In terms of FDG analysis, the difference between direct speech and indirect speech is shown to correspond to a difference in the type of entity denoted by the speech complement (a discourse act versus communicated content ). The ±frame distinction is reflected in the representation of the communicated content (fully expressed versus holophrastic). The even less prototypical constructions, too, are given their own analysis; it is demonstrated that these constructions occupy their own position in the matrix.

This article analyses the interaction between semantic and pragmatic information in illocution and focus marking in Umpithamu (Cape York, Australia), and it uses these data to argue that while FDG can deal with the basics of the semantics-pragmatics interface, it lacks three crucial features to represent the interaction between semantic and pragmatic information in an adequate way. The marking of illocution illustrates the importance of preferred interpretations that are halfway between encoded and incidental aspects of interpretation, driven by general pragmatic principles. While such preferred interpretations are crucial from the perspective of diachrony and language processing, they do not have a clear place in the FDG model. The marking of focus illustrates the existence of categories in the case system that combine interpersonal and representational aspects of organization in a nonarbitrary way, which can form a diachronic pathway between the interpersonal and the representational levels. While the synchronically dual nature of categories is easy to represent in FDG, the nonarbitrary link between the two functions, and its potential diachronic consequences, do not have an obvious place in the model.

Any description of Scottish Gaelic must make continual reference to interpersonal factors, since the occupancy of the first position in the clause (P I ) is codetermined by the illocutionary status of the discourse act being carried out through the formulation of that clause. Gaelic is a language in which syntactic functions play no part, but which gives great prominence to pragmatic functions. Although topic has no role in Gaelic grammar, focus and contrast both impinge upon its morphology and syntax, with the cleft construction extending from being an indicator of contrast to a marker of focus, where no contrast is intended.

Algonquian languages mark discourse topicality on nouns and verbs by means of obviation morphology: the proximate marks topical participants, while the obviative marks nontopical participants. In Blackfoot, portmanteau obviation morphemes express a combination of pragmatic and semantic distinctions, including illocution, referentiality, topicality, animacy, person and number. An analysis of the proximate as the expression of grammaticalized topic within the framework of functional discourse grammar accounts for the distribution of proximates and obviatives in a narrative text. In possessive constructions, the possessor may be proximate or obviative, while the possessed must be obviative. This restriction is explained in terms of a hierarchy alignment requirement that forbids constructions in which nonalignment is not morphologically marked. Proximate assignment to a new entity is called a proximate shift; it signals the beginning of a new topic chain. Certain complement clauses do not allow proximate shifts: in such clauses, the proximate entity must be coreferential with the proximate entity of the matrix clause. This restriction is accounted for by assuming that such clauses do not constitute independent discourse acts. They share a pragmatic and deictic center with their matrix clause, and therefore do not allow independent topic assignment, resulting in obligatory coreferentiality of matrix and complement clause topic.

The relationship between question words (Q-words) and focus is far more complex than generally assumed, in FDG or elsewhere. Q-words are not always contrastive, the focus of a Q-word clause can have more elements than the Q-word in its scope, and Q-word clauses may contain another, separate focus besides the Q-word. This becomes apparent when studying languages in which the Q-word seems to be freely distributed across the clause. A clear example is Hittite, an extinct Indo-European SOV language in which the placement of Q-words is not restricted to the preverbal focus position expected for SOV languages. In this article I will show that Hittite uses multiple strategies for the placement of Q-words, depending on the degree of counterexpectancy of the Q-word question. This motivation for the placement of the (always focused) Q-constituent is shared with the placement of focus constituents in Hittite declaratives. We should thus abandon the notion of a special question focus type, and describe the focus structure at the interpersonal level in FDG irrespective of the type of illocutionary act. As a corollary, the existential presupposition of a Q-word question, often taken to be its only pragmatic presupposition, is detached from its focus structure. In this article the existential presupposition is considered an inherent property of the Q-word, not of the question. Another consequence is that an FG-type focus typology as already established for declaratives can be set up for Q-word questions as well.

Rembarrnga (Gunwinyguan, Australian) is a polysynthetic language exhibiting complex verbal morphology with multiple (typically obligatory) slots in the verbal template for the expression and discourse manipulation of core propositional actants. External to the verbal word there exist a number of syntactic slots which may function to identify or further specify event participants. Through the investigation of the distinct function of a number of formally alternative means available to a speaker for tracking discourse participants in Rembarrnga, this article is intended to show the subtle interaction between grammatical properties and discourse strategies of the language. These interactions are to some extent driven by the interpersonal level of grammar in which a speaker signifies to an addressee through grammatical means outlined here the assumptions as to what discourse status the speaker wishes — and believes — the addressee will attribute to discourse participants; as tacit (e.g., presupposed or given) or explicit (e.g., new). The discussion herein also proposes an appropriate FDG representation of these structures at the interpersonal level of grammar.

Many languages employ articles as an expression device to mark identifiability. There are, however, other means to mark identifiability in languages. In this paper I show that one of the two sets of cross-referencing markers on the transitive verb in Hungarian serves this purpose, i.e., it represents the instruction from the speaker that he considers the referent of the object argument to be identifiable to the hearer. Next, attention is paid to four sets of Hungarian data which are seemingly at odds with this claim. These sets relate to the use of particular person distinctions, indefinite noun phrases which contain either a possessive restrictor or the interrogative pronoun melyik ‘which’, and to relative clauses. Apart from a descriptive account of the relevant data, the paper also offers a theoretical account within the framework of functional discourse grammar, where the speaker assessment of identifiability of referents is formally represented by operators which apply to the referential act, represented by the variable R at the interpersonal level. Distinctions made at the interpersonal level thus affect the choice of verbal forms in Hungarian.

This article gives a diachronic account of adnominal modification from Proto-Indo-European to present-day Dutch. The main conclusion is that through the ages, noun phrases appear to “fold out”: they acquire their layered structure for different lexical modifiers over time. The latest stage in this process is the fairly recent development of a specific slot for interpersonal modification of the whole noun phrase. The different stages in the diachronic development are described with the layered and modular representation of functional discourse grammar (FDG).