Syntax-discourse interface

In keeping with our observation about the increasing mastery of DGS properties that involve the interface between syntax and discourse the analysis of file 3 also reveals a skilful use of those mechanisms that contribute to the creation of cohesion.

Referential establishment and maintenance. In file 3, Simon uses several means to establish and maintain reference. For further illustration consider (166). In (166a), the first clause of this narrative sequence, Simon uses the body as a classifier in an SRF to express that the boy supports himself on something, leaning forward, before he finally spots the frog. In (166b) the boy’s position and his discovery are expressed simultaneously through a complex classifier construction: the h2-classifier is used to background the information that the boy is leaning on something and the dominant hand is used to foreground the information about the boy spotting the frog through the agreement verb spot. Note that the locus encoding the object argument in this verb and the locus associated with [det_exist]_g used in (166c) to inform about the frog’s location coincide, which illustrates not only the diversity of linguistic means used to mark agreement, but also the mastery of the mechanisms necessary to mark reference maintenance.

For further illustration of how loci are established and maintained we might consider example (167). The locus for the frog is established through the agreement verb look in (167a); referential identity is marked through the choice of the same locus for the frog in the next clause (167b), in which the auxiliary pam marks the agreement between the subject (the dog and the boy) and the object (the frog)

(note, however, that the word order in (167b) follows the VO pattern which could be an instance of structural borrowing of German; recall that we remarked upon this phenomenon in our discussion of Muhammed’s data). det_loc in (167 c) associates a locus with the location of the frog, sitting inside a jar.

At times, however, referential identity is difficult to establish in this file. Ambiguities obtain where Simon does not establish referential loci contrastively, or where he does not use overt lexical expressions to reintroduce the boy as a protagonist. Consider, for example, the two successive sequences in (168) and (169), involving the dog and the boy respectively as a protagonist. In (168) the referential framework is shifted in a quotation environment, in which the dog says that he doesn’t see anything (because his head is stuck in the jar). Notice that the POV is signalled through the matrix verb say, a change of facial expression, and a shift of body orientation and eye gaze direction to the left. Example (168) is followed by the sequence in (169), in which the boy is reintroduced as a protagonist. The subject is dropped in (169a), the utterance that precedes a POV in (169b), in which the signer adopts the perspective of the boy. Notice that the non-manual means used to signal this shift in (169b) coincide with the ones used to mark the shifted perspective in (168), indicating that no distinction is made at this point regarding the loci associated with the two referents (the dog and the boy). It must be noted that Simon uses of a full NP at the beginning of the narrative passage involving the dog as a protagonist, which contributes to an unambiguous interpretation of referential identity in (168). This is not the case in (169a), in which the boy is reintroduced as a protagonist. So, in this narrative, too, it seems the boy is chosen as a thematic subject, an observation we will take up below when we discuss the choice of reference forms used to refer to the story characters. The same phenom?enon can be observed in the sequence describing the boy’s falling on the deer, after his misperception of the antlers. Notice that the NP deer in (170a) signals the referential shift, as the signer adopts the perspective of the deer, marked non-manually by a slight change in body orientation and eye gaze direction (to the right). Again, the change of perspective in (170e) recounting the boy’s falling from the deer’s head is not signalled lexically.

Finally, another problematic sequence involving a shift of perspective is provided in (171). Notice that Simon has not introduced the agents of the activity before. The audience learns that the boy has found the frog, but is not informed about the family that is together with him, including the parents that are commonly identified as the ones offering the boy one of their offspring. The referential shift is marked non-manually through a change in body orientation and eye gaze direction (to the right). Previously, Simon used this locus when he adopted the perspective of the boy. So, although the object is clear, the subject remains unclear, leaving the audience to infer its identity.

Reference forms and functions. Turning to the choice of reference forms used to refer to story characters in file 3, it is interesting to note that reference to the boy only occurs once through an NP in the introductory statement at the beginning of the narration and shortly after, when Simon explains that the boy wants to go to sleep because he is tired. All other references to the boy throughout the narrative occur without an overt lexical expression, which, as we remarked upon previously, makes it difficult at times to appropriately determine referential identity in some sequences. This phenomenon is reflected in the proportion of subject-drop in reintroductory contexts; notice that the percentage, though lower than in file 1, remains relatively high. Indicentally, the percentage of 10.7% out of a total of 23.2% of forms serving the function of reintroduction is the same for NPs (cf. also Figure 3.4).

Table 3.23: Reference forms and functions in Simon’s file 3.*

*Expressed as a percentage of the total number of reference forms (proportions of forms used for respective function in brackets). Absolute numbers are provided in the Appendix Table C-4.

Figure 3.4:

Proportion of reference forms and functions in Simon’s file 3.

The assumption that the boy is chosen as a thematic subject is corroborated by the choice of reference forms to refer to the other story characters. Reintroduction of the dog as a protagonist always occurs via an NP. Two other characters, the mole and the deer, only appear in individual scenes respectively. The frog, in turn, is introduced via an NP at the beginning and explicitly referred to whenever the boy calls for him or asks another character about his whereabouts. Whenever the result of a search in a certain place turns out to be negative, however, the expression “not there” does not contain an explicit reference to the runaway frog. This phenomenon, as we explained before, might be an effect of the main story topic (the search of the runaway frog). For further illustration consider example (172), which documents subject drop in a sequence, in which the boy is reintroduced as a protagonist (following the scene describing the frog’s escape). Also, it is not mentioned explicitly in (172c) that it is the frog that is absent. While referential identity of the subject in (172a) and (172b) must be inferred from the context, the identity of the object arguments in (172b) and (172d) is clear because verb forms pick up the locus associated previously with the frog.

Expression of spatial relations. Table 3.24 provides an overview of the linguistic devices used by Simon in file 3 to express figure-ground relations. As we can see, information on the ground is always provided in this file, and it is always expressed clearly. This observation marks an important difference to file 1. For further illustration, consider the sequences in (173) and (174). In example (173) the boy is reported to look at and then climb on a rock. The rock is introduced via an NP followed by a specification of its shape. Notice that in (174), Simon first reports the falling of the boy (174a), which is followed by a sequence with the information on where the boy falls down (into the water) (cf. (174d)).

Table 3.24: Expression of figure-ground relations in Simon’s file 3 narrative.*

* The nature of the location is specified a posteriori (cf. (167)).

Language contact

Candidates for language contact phenomena include constructions with pam involving an erroneous word order, as well as utterances involving combinations of a predicate with the auxiliary verb has, an LBG sign (cf. (175)) or the combination of the verb make with the verb ask (cf. (176c)). These expressions do not correspond with German expressions (in German, the equivalent of (175) would involve the auxiliary sein (‘to be’) and the equivalent of (176c) the noun-verb combination Frage stellen, ‘ask a question’, if not the use of the verb fragen, ‘to ask’). Again, it seems the use of LBG elements results in constructions that are neither DGS nor German.