Reference forms and functions

The skilful organisation of referential content in a narrative involves the syntax-discourse (or pragmatics) interface. The acquisition of the mechanisms needed to create cohesion and coherence in sign language discourse, much like in spoken language discourse, comprises mastery of the different types of reference forms, the syntactic context they may appear in, as well as the functions they fulfil in a given discourse context. As pointed out by Verbist (2010: 116) “children not only need to learn to generate different nominal expressions in the appropriate positions at the syntactic level, but also to identify the discourse information status.”

While young (hearing) infants have been found to master the new-given information distinction (by the age of two, or even younger, as even at the one word stage, the words produced tend to express new information) (cf. Verbist 2010: 116), “[m]apping this pragmatic knowledge on to the syntactic set of co-referring expressions is a much more complex operation.” Indeed, form-function relations in learners’ productions have been found to change over time, which is reflected also in the use of different strategies in the organisation of a narrative.

Morgan (2000: 282), who draws on the work of Karmiloff-Smith (1985), summarises the developmental phases distinguished in the literature with respect to the strategies learners use in their organisation of narratives. Children before age 5 begin with a bottom-up strategy, whereby the focus is put on the local sentential level (and, hence, on the relations of referents within a sentence). Where picture books are used as in the elicitation of narratives, the organisation of the narrative produced follows the order of these pictures. In the next phase, after age 5, children choose a “thematic subject” for their story (the main protagonist) following a top-down strategy in the narrative organisation. Finally, older children and adults have been found to make a balanced use of top-down and bottom-up strategies. Narratives are organised in relation to larger discourse units.

Adultlike usage of this pragmatic knowledge continues to develop in the teenage years. It has been observed, for example, that the expression of narrative parts involving two characters in separate but co-occurring activities, is managed only by the 11-13 year old participants in a study of children age 4-13 acquiring BSL (Morgan 2006). Only this age group managed to refer to two characters involved in the same episode and to overlay perspectives through the sequential and simultaneous use of fixed and shifted referential frameworks (for similar observations concerning the use of perspective in learners of ASL and NGT, see Slobin et al. 2003: 291f.).

Reference forms and functions. Choice of reference forms and the functions they fulfil at the discourse level have been investigated in a comparative study of deaf children’s and adults’ narrative productions in BSL, elicited on the basis of the famous frog story (Morgan 2006). As outlined in section 3.1.4.8 three referential functions reference forms might serve in a discourse context are commonly distinguished, namely, reference introduction, reintroduction, and maintenance.

As for the form-function patterns used by the adult participants in the study, Morgan’s analysis (2006: 325) revealed that reference maintenance was primarily expressed through role shift (about 59%), followed by entity classifiers (31%), and noun phrases (6%). As for the introduction of new referents, Morgan remarks that this seldom occurs through entity classifiers (only 4%, used with a cataphoric ref?erence, that is, these forms were immediately followed by a noun phrase identifying the referent explicitly), as new referents were typically introduced via noun phrases (referential shift was not found to be used for introduction). Turning to the data of the children participating in the study (aged 4-13), Morgan (2006: 324f.) remarks on the progressive decrease of ambiguous forms (that is those forms for which it was impossible to identify the character they referred to), from 16% at age 4 to 0.2% at age 13.

As for the referential functions served by the different reference forms the study revealed the following:

• (a) Noun phrases. In all age groups, noun phrases are used to introduce and reintroduce characters, although the youngest children (age 4-6) often failed to use noun phrases to introduce new characters. The highest percentage of use noun phrases for reference maintenance was observed in the younger children (22,5%) (a high percentage reappears at age 11, which is interpreted as an indication for the continuing narrative development). This use of noun phrases is interpreted to reflect the learners’ focus on reference at the sentential level (Morgan 2006). Interestingly, the use of noun phrases for reference maintenance equals the first phase identified in hearing children learning a spoken language.
• (b) Entity classifiers. Entity classifiers were found to be used less by younger participants for reference maintenance (about 12,5%) than by adults (about 31%). But the younger children used this form to introduce referents (8%), without a cataphoric or following noun phrase (as adults would do). Hence, children master the use of these forms at sentence level, but the narrative knowledge about the pragmatic functions is still lacking. Interestingly, in the youngest children’s productions classifiers appeared across the three referential functions fairly uniformly.
• (c) Referential shift. Finally, regarding referential shift, it was found to be predominantly used for reference maintenance across all age groups. Youngest children (age 4-6), however, use this form for about 11,25% to introduce new referents, which can be interpreted as one of the main causes for the referential ambiguity observed in their stories.

Morgan (2006: 327) concludes that “[i]n general, control of the pragmatic role of entity classifiers and role shift in discourse develops gradually with initial mastery at the sentential level, where young children may use these constructions correctly but fail to use them appropriately in relation to their new referential functions in discourse.” Crucially, the increasing mastery of the functions served by reference forms at the narrative level reflect “a major growth in the child’s pragmatic abilities to assess the knowledge of the listener as well as monitor the narrative for ambiguity” (Morgan 2006: 323).